diff --git a/.github/workflows/ci.yml b/.github/workflows/ci.yml
index 0402ea0e..fb319d44 100644
--- a/.github/workflows/ci.yml
+++ b/.github/workflows/ci.yml
@@ -47,13 +47,14 @@ jobs:
           pytest tests/test_alert_checker.py tests/test_schedule_runner.py tests/test_export_audit.py \
             tests/test_export_audit_coverage.py tests/test_audit_tools.py tests/test_audit_tools_expanded.py \
             tests/test_audit_tools_coverage.py tests/test_audit_tools_dispatch_coverage.py \
-            tests/test_export_custom_coverage.py tests/test_export_artifacts_coverage.py \
-            tests/test_export_compare_coverage.py tests/test_export_tools_coverage.py \
-            tests/test_image_tools.py tests/test_export_custom.py tests/test_export_artifacts.py \
-            tests/test_export_compare.py \
-            tests/test_mcp_registry.py tests/test_mcp_resources.py \
+            tests/test_audit_tools_links_extras.py tests/test_export_custom_coverage.py \
+            tests/test_export_artifacts_coverage.py tests/test_export_compare_coverage.py \
+            tests/test_export_tools_coverage.py tests/test_image_tools.py tests/test_export_custom.py \
+            tests/test_export_artifacts.py tests/test_export_compare.py tests/test_export_workbook.py \
+            tests/test_export_sitemap.py tests/test_mcp_registry.py tests/test_mcp_resources.py \
+            tests/test_tools_branch_coverage.py \
             --cov=website_profiling.tools --cov-config=.coveragerc.tools \
-            --cov-report=term-missing --cov-fail-under=95 -q -o addopts=
+            --cov-report=term-missing --cov-fail-under=100 -q -o addopts=
       - name: CLI smoke
         run: python -m src --help
 
diff --git a/alembic/versions/013_crawl_discovery_and_link_edges.py b/alembic/versions/013_crawl_discovery_and_link_edges.py
new file mode 100644
index 00000000..a666036c
--- /dev/null
+++ b/alembic/versions/013_crawl_discovery_and_link_edges.py
@@ -0,0 +1,56 @@
+"""Crawl discovery mode on crawl_runs and rich link_edges table.
+
+Revision ID: 013_crawl_discovery_edges
+Revises: 012_chat_sessions
+"""
+from __future__ import annotations
+
+from alembic import op
+
+revision = "013_crawl_discovery_edges"
+down_revision = "012_chat_sessions"
+branch_labels = None
+depends_on = None
+
+
+def upgrade() -> None:
+    op.execute("""
+        ALTER TABLE crawl_runs
+            ADD COLUMN IF NOT EXISTS discovery_mode TEXT DEFAULT 'spider';
+
+        CREATE TABLE IF NOT EXISTS link_edges (
+            crawl_run_id BIGINT NOT NULL REFERENCES crawl_runs(id) ON DELETE CASCADE,
+            from_url TEXT NOT NULL,
+            to_url TEXT NOT NULL,
+            anchor_text TEXT NOT NULL DEFAULT '',
+            rel TEXT NOT NULL DEFAULT '',
+            is_nofollow BOOLEAN NOT NULL DEFAULT FALSE,
+            is_sponsored BOOLEAN NOT NULL DEFAULT FALSE,
+            is_ugc BOOLEAN NOT NULL DEFAULT FALSE,
+            link_type TEXT NOT NULL DEFAULT 'internal',
+            PRIMARY KEY (crawl_run_id, from_url, to_url, anchor_text, rel)
+        );
+        CREATE INDEX IF NOT EXISTS idx_link_edges_run_from
+            ON link_edges(crawl_run_id, from_url);
+        CREATE INDEX IF NOT EXISTS idx_link_edges_run_to
+            ON link_edges(crawl_run_id, to_url);
+
+        CREATE TABLE IF NOT EXISTS saved_crawl_filters (
+            id BIGINT GENERATED BY DEFAULT AS IDENTITY PRIMARY KEY,
+            property_id BIGINT NOT NULL REFERENCES properties(id) ON DELETE CASCADE,
+            name TEXT NOT NULL,
+            filter_json JSONB NOT NULL DEFAULT '{}',
+            created_at TIMESTAMPTZ NOT NULL DEFAULT now(),
+            UNIQUE (property_id, name)
+        );
+        CREATE INDEX IF NOT EXISTS idx_saved_crawl_filters_property
+            ON saved_crawl_filters(property_id);
+    """)
+
+
+def downgrade() -> None:
+    op.execute("""
+        DROP TABLE IF EXISTS saved_crawl_filters CASCADE;
+        DROP TABLE IF EXISTS link_edges CASCADE;
+        ALTER TABLE crawl_runs DROP COLUMN IF EXISTS discovery_mode;
+    """)
diff --git a/docs/GLOSSARY.md b/docs/GLOSSARY.md
index bc8c9ca0..436850f4 100644
--- a/docs/GLOSSARY.md
+++ b/docs/GLOSSARY.md
@@ -39,17 +39,21 @@ UI terms agencies recognize, mapped to internal keys and data sources.
 | Scheduled audits | `properties.schedule_cron`, `/api/schedule/check` | Cron + pipeline spawn | Recurring site audit — see [OPS.md](OPS.md) |
 | Property alerts | `alert_webhook_url`, `/api/alerts/check` | Health snapshot rules | Ops notifications |
 | Content brief | Keywords Brief button, `/api/keywords/content-brief` | LLM or deterministic | Content planning |
-| AI issue fix | `llm_recommendation`, `/api/issues/fix-suggestion` | LLM on demand + report build | Actionable remediation |
+| AI fix suggestions | `llm_recommendation`, `/api/ai/fix-suggestion`, `/api/issues/fix-suggestion` (legacy) | LLM on demand + report build | Actionable remediation across Issues, Lighthouse, Security, and other surfaces |
 | AI Chat | `/chat`, `/api/chat`, `chat_sessions` | LLM + read-only audit tools | Conversational site audit queries |
 | MCP tools | `python -m website_profiling.mcp` | Same `audit_tools` as chat | Cursor / Claude Desktop integration — see [MCP.md](MCP.md) |
 | Read-only session | `AUTH_DEFAULT_ROLE=client-readonly`, `/api/auth/session` | Session cookie | Client view-only access |
 | Export executive summary | `export_audit_html/pdf/csv`, `export_audit_report` (chat/MCP), Export view | Report payload + optional AI | Client deliverable |
+| ads.txt / security.txt | `site_level`, `get_ads_txt_status`, `get_security_txt_status` | Root file fetch at report build | Publisher / contact file hygiene |
+| Subdomain inventory | `subdomains`, `list_subdomains`, `/subdomains` view | Crawl + GSC + optional crt.sh | Host footprint vs crawl scope |
+| Contact intelligence | `contact_intelligence`, `get_contact_intelligence`, `/contacts` view | Crawl schema/mailto + security.txt + RDAP org | Business identity consistency |
 
 ## Metric names
 
 | UI | Field | Source |
 |----|-------|--------|
-| Inlinks | `inlinks` | Crawl graph |
+| Impact score | `impact_score` on issues | GSC clicks + GA4 sessions + priority weight (see below) |
+| Link edges | `link_edges`, `link_rel_summary` | Crawl anchor/rel attributes |
 | Outlinks | `outlinks` | Crawl graph |
 | Status code | `status` | HTTP |
 | Crawl rendering | `crawl_render_mode` on run; `fetch_method` per URL | `static`, `javascript`, or `auto` crawl config; `static` vs `rendered` per page |
@@ -62,6 +66,8 @@ UI terms agencies recognize, mapped to internal keys and data sources.
 | On-site frequency | `volume` (heuristic) | Estimated from crawl |
 | Sessions | GA4 metrics | Analytics |
 
+**Impact score:** `priority_weight + (gsc_clicks × 10) + (ga4_sessions × 5)` with Critical=1000, High=100, Medium=10, Low=1.
+
 ## Provenance badges
 
 | Badge | Meaning |
diff --git a/docs/MCP.md b/docs/MCP.md
index dd473f8e..d3ba6f94 100644
--- a/docs/MCP.md
+++ b/docs/MCP.md
@@ -43,11 +43,11 @@ Add to `.cursor/mcp.json` (or Cursor MCP settings):
 | `audit://glossary` | Excerpt from `docs/GLOSSARY.md` |
 | `audit://tools` | Tool catalog grouped by SEO domain |
 
-## Tools (171 read-only + export)
+## Tools (176 read-only + export)
 
 ### Export and deliverables
 
-`export_audit_report`, `export_compare_csv`, `export_list_as_csv`, `compose_custom_report`, `export_custom_report`, `list_export_formats`
+`export_audit_report`, `export_compare_csv`, `export_list_as_csv`, `export_sitemap_xml`, `validate_rich_results`, `compose_custom_report`, `export_custom_report`, `list_export_formats`
 
 Full audit exports reuse the same generators as the Export view (PDF requires `reportlab`). Export tools store files as artifacts (24h TTL); in-app chat renders download buttons via `/api/chat/artifacts/{id}`.
 
@@ -79,7 +79,7 @@ Size-based tools require `probe_image_inventory=true` in pipeline config when bu
 
 ### Links and architecture
 
-`list_orphan_pages`, `get_top_linked_pages`, `get_top_crawled_pages`, `get_outbound_link_domains`, `get_link_graph_summary`, `get_url_fingerprints`, `list_broken_link_sources`, `get_mime_type_breakdown`, `get_title_length_distribution`, `get_domain_link_distribution`, `get_outlink_distribution`
+`get_link_rel_summary`, `get_inlink_anchors`, `list_nofollow_internal_links`, `list_orphan_pages`, `get_top_linked_pages`, `get_top_crawled_pages`, `get_outbound_link_domains`, `get_link_graph_summary`, `get_url_fingerprints`, `list_broken_link_sources`, `get_mime_type_breakdown`, `get_title_length_distribution`, `get_domain_link_distribution`, `get_outlink_distribution`
 
 ### Indexation and international
 
@@ -115,12 +115,13 @@ Size-based tools require `probe_image_inventory=true` in pipeline config when bu
 
 ## Future pipeline items (not yet exposed as tools)
 
-These require additional crawl or third-party integrations before dedicated tools are useful:
+These require additional third-party integrations or product scope beyond current crawl data:
 
-- Google Rich Results / schema validation API
-- Full backlink index and anchor-text analytics
-- axe / color-contrast accessibility audits
+- Full backlink index and anchor-text analytics (beyond GSC Links import)
 - SERP rank tracking beyond GSC position snapshots
+- Standalone Google Rich Results Test API (current `validate_rich_results` uses crawl heuristics + GSC URL Inspection)
+
+Already available: `validate_rich_results`, `export_sitemap_xml`, workbook export, axe audits via `enable_axe` on browser crawls.
 
 ## Example prompts
 
diff --git a/input.txt.example b/input.txt.example
index 4d1b4d0e..a6a9c56e 100644
--- a/input.txt.example
+++ b/input.txt.example
@@ -18,6 +18,16 @@ content_excerpt_max_chars = 4096
 preserve_crawl_history = true
 crawl_stream_to_db = false
 crawl_exclude_urls = 
+crawl_discovery_mode = spider
+crawl_url_list = 
+crawl_user_agent_preset = default
+crawl_user_agent_custom = 
+crawl_auth_username = 
+crawl_auth_password = 
+crawl_extra_headers = 
+crawl_cookies = 
+crawl_robots_txt_override = 
+custom_extractors = 
 # crawl_render_mode: static | javascript | auto (auto = static first, browser when SPA heuristics match)
 crawl_render_mode = static
 crawl_js_concurrency = 3
@@ -46,6 +56,9 @@ max_image_probe_urls = 500
 image_probe_concurrency = 6
 image_probe_timeout = 8
 image_unoptimized_min_kb = 200
+enable_subdomain_discovery = true
+subdomain_ct_lookup = true
+enable_rdap_org_lookup = true
 
 # --- Lighthouse ---
 lighthouse_url = 
@@ -56,6 +69,13 @@ lighthouse_iterations = 1
 run_lighthouse = true
 run_lighthouse_on_pages = true
 enable_crux = false
+enable_rich_results_validation = false
+google_rich_results_api_key =
+enable_axe = false
+enable_spell_check = false
+enable_html_validation = false
+enable_amp_audit = false
+enable_wayback_lookup = false
 competitor_domains =
 bing_webmaster_api_key =
 serp_api_key =
diff --git a/pipeline-config.example.txt b/pipeline-config.example.txt
index b1c384b3..a43055d3 100644
--- a/pipeline-config.example.txt
+++ b/pipeline-config.example.txt
@@ -19,6 +19,16 @@ content_excerpt_max_chars = 4096
 preserve_crawl_history = true
 crawl_stream_to_db = false
 crawl_exclude_urls = 
+crawl_discovery_mode = spider
+crawl_url_list = 
+crawl_user_agent_preset = default
+crawl_user_agent_custom = 
+crawl_auth_username = 
+crawl_auth_password = 
+crawl_extra_headers = 
+crawl_cookies = 
+crawl_robots_txt_override = 
+custom_extractors = 
 # crawl_render_mode: static | javascript | auto
 crawl_render_mode = static
 crawl_js_concurrency = 3
@@ -47,6 +57,9 @@ max_image_probe_urls = 500
 image_probe_concurrency = 6
 image_probe_timeout = 8
 image_unoptimized_min_kb = 200
+enable_subdomain_discovery = true
+subdomain_ct_lookup = true
+enable_rdap_org_lookup = true
 
 # --- Lighthouse ---
 lighthouse_url = 
@@ -57,6 +70,13 @@ lighthouse_iterations = 1
 run_lighthouse = true
 run_lighthouse_on_pages = true
 enable_crux = false
+enable_rich_results_validation = false
+google_rich_results_api_key =
+enable_axe = false
+enable_spell_check = false
+enable_html_validation = false
+enable_amp_audit = false
+enable_wayback_lookup = false
 competitor_domains =
 bing_webmaster_api_key =
 serp_api_key =
diff --git a/requirements-optional.txt b/requirements-optional.txt
new file mode 100644
index 00000000..488a3046
--- /dev/null
+++ b/requirements-optional.txt
@@ -0,0 +1,3 @@
+# Optional audit dependencies (install when enabling spell-check / HTML validation extras)
+pyspellchecker>=0.8.1
+html5lib>=1.1
diff --git a/scripts/local-test.ps1 b/scripts/local-test.ps1
index a354c8fe..14b0e56f 100644
--- a/scripts/local-test.ps1
+++ b/scripts/local-test.ps1
@@ -245,7 +245,7 @@ function Invoke-PytestReporting {
 }
 
 function Invoke-PytestTools {
-    Write-Log "Pytest (tools coverage gate, 95%)"
+    Write-Log "Pytest (tools coverage gate, 100%)"
     & $VENV_PYTEST `
         tests/test_alert_checker.py `
         tests/test_schedule_runner.py `
@@ -265,10 +265,11 @@ function Invoke-PytestTools {
         tests/test_export_compare.py `
         tests/test_mcp_registry.py `
         tests/test_mcp_resources.py `
+        tests/test_tools_branch_coverage.py `
         --cov=website_profiling.tools `
         --cov-config=.coveragerc.tools `
         --cov-report=term-missing `
-        --cov-fail-under=95 `
+        --cov-fail-under=100 `
         -q `
         -o addopts=
     Assert-LastExitCode "Tools coverage gate failed"
diff --git a/scripts/local-test.sh b/scripts/local-test.sh
index 960683fb..a24022a8 100755
--- a/scripts/local-test.sh
+++ b/scripts/local-test.sh
@@ -119,10 +119,14 @@ run_pytest_reporting() {
     tests/test_categories_roadmap.py \
     tests/test_report_categories_golden.py \
     tests/test_categories_coverage.py \
+    tests/test_contrast_issues.py \
     tests/test_indexation_coverage.py \
     tests/test_crawl_segments.py \
     tests/test_terminology.py \
     tests/test_compare_payload.py \
+    tests/test_optional_audits.py \
+    tests/test_property_profile.py \
+    tests/test_reporting_gaps.py \
     --cov=website_profiling.reporting \
     --cov-config=.coveragerc.reporting \
     --cov-report=term-missing \
@@ -133,7 +137,7 @@ run_pytest_reporting() {
 
 run_pytest_tools() {
   [[ "$PYTEST_NO_COV" -eq 1 ]] && return 0
-  log "Pytest (tools coverage gate, 95%)"
+  log "Pytest (tools coverage gate, 100%)"
   "$VENV/bin/pytest" \
     tests/test_alert_checker.py \
     tests/test_schedule_runner.py \
@@ -143,6 +147,7 @@ run_pytest_tools() {
     tests/test_audit_tools_expanded.py \
     tests/test_audit_tools_coverage.py \
     tests/test_audit_tools_dispatch_coverage.py \
+    tests/test_audit_tools_links_extras.py \
     tests/test_export_custom_coverage.py \
     tests/test_export_artifacts_coverage.py \
     tests/test_export_compare_coverage.py \
@@ -151,12 +156,15 @@ run_pytest_tools() {
     tests/test_export_custom.py \
     tests/test_export_artifacts.py \
     tests/test_export_compare.py \
+    tests/test_export_workbook.py \
+    tests/test_export_sitemap.py \
     tests/test_mcp_registry.py \
     tests/test_mcp_resources.py \
+    tests/test_tools_branch_coverage.py \
     --cov=website_profiling.tools \
     --cov-config=.coveragerc.tools \
     --cov-report=term-missing \
-    --cov-fail-under=95 \
+    --cov-fail-under=100 \
     -q \
     -o addopts=
 }
diff --git a/src/website_profiling/analysis/page.py b/src/website_profiling/analysis/page.py
index 6ae3612e..e9185a97 100644
--- a/src/website_profiling/analysis/page.py
+++ b/src/website_profiling/analysis/page.py
@@ -85,6 +85,97 @@ def walk(obj: object) -> bool:
     return walk(data)
 
 
+_CONTACT_ORG_TYPES = frozenset({
+    "organization",
+    "localbusiness",
+    "corporation",
+    "store",
+    "restaurant",
+    "professionalService",
+    "newsmediaorganization",
+})
+_CONTACT_CAP = 10
+
+
+def _normalize_type_name(raw: object) -> str:
+    if isinstance(raw, str):
+        return raw.strip()
+    if isinstance(raw, list) and raw:
+        return _normalize_type_name(raw[0])
+    return ""
+
+
+def _collect_json_ld_types(data: object, types: set[str]) -> None:
+    if isinstance(data, dict):
+        t = data.get("@type")
+        name = _normalize_type_name(t)
+        if name:
+            types.add(name)
+        graph = data.get("@graph")
+        if isinstance(graph, list):
+            for item in graph:
+                _collect_json_ld_types(item, types)
+        for key, val in data.items():
+            if key in ("@graph", "@type"):
+                continue
+            if isinstance(val, (dict, list)):
+                _collect_json_ld_types(val, types)
+    elif isinstance(data, list):
+        for item in data:
+            _collect_json_ld_types(item, types)
+
+
+def _format_postal_address(addr: object) -> str:
+    if isinstance(addr, str):
+        return addr.strip()[:500]
+    if not isinstance(addr, dict):
+        return ""
+    parts: list[str] = []
+    for key in ("streetAddress", "addressLocality", "addressRegion", "postalCode", "addressCountry"):
+        val = addr.get(key)
+        if isinstance(val, str) and val.strip():
+            parts.append(val.strip())
+    return ", ".join(parts)[:500]
+
+
+def _append_contact(signals: dict[str, list[str]], key: str, value: object) -> None:
+    if not isinstance(value, str):
+        return
+    val = value.strip()
+    if not val:
+        return
+    bucket = signals.setdefault(key, [])
+    if len(bucket) >= _CONTACT_CAP:
+        return
+    if val not in bucket:
+        bucket.append(val)
+
+
+def _collect_json_ld_contacts(data: object, signals: dict[str, list[str]]) -> None:
+    if isinstance(data, dict):
+        type_name = _normalize_type_name(data.get("@type")).lower()
+        is_org = any(t in type_name for t in _CONTACT_ORG_TYPES) or type_name in _CONTACT_ORG_TYPES
+        if is_org:
+            _append_contact(signals, "organization_names", data.get("name"))
+            _append_contact(signals, "emails", data.get("email"))
+            _append_contact(signals, "phones", data.get("telephone"))
+            addr = _format_postal_address(data.get("address"))
+            if addr:
+                _append_contact(signals, "addresses", addr)
+        graph = data.get("@graph")
+        if isinstance(graph, list):
+            for item in graph:
+                _collect_json_ld_contacts(item, signals)
+        for key, val in data.items():
+            if key in ("@graph",):
+                continue
+            if isinstance(val, (dict, list)):
+                _collect_json_ld_contacts(val, signals)
+    elif isinstance(data, list):
+        for item in data:
+            _collect_json_ld_contacts(item, signals)
+
+
 def analyze_html(
     html: str,
     page_url: str,
@@ -322,6 +413,14 @@ def _script_in_head(sc) -> bool:
             image_urls.append(url)
     out["image_urls"] = _cap(image_urls)
 
+    json_ld_types: set[str] = set()
+    contact_signals: dict[str, list[str]] = {
+        "emails": [],
+        "phones": [],
+        "addresses": [],
+        "organization_names": [],
+    }
+
     # JSON-LD
     for idx, sc in enumerate(soup.find_all("script", type=lambda t: t and "ld+json" in str(t).lower())):
         raw = (sc.string or "").strip()
@@ -332,6 +431,8 @@ def _script_in_head(sc) -> bool:
         except json.JSONDecodeError:
             warn("json_ld_parse", "medium", "Invalid JSON-LD block", f"Block index {idx}")
             continue
+        _collect_json_ld_types(data, json_ld_types)
+        _collect_json_ld_contacts(data, contact_signals)
         if _json_ld_missing_type(data):
             warn(
                 "json_ld_missing_type",
@@ -341,6 +442,20 @@ def _script_in_head(sc) -> bool:
             )
             break
 
+    if json_ld_types:
+        out["json_ld_types"] = sorted(json_ld_types)[:_CONTACT_CAP]
+
+    for a in soup.find_all("a", href=True):
+        href = (a.get("href") or "").strip()
+        lower = href.lower()
+        if lower.startswith("mailto:"):
+            _append_contact(contact_signals, "emails", href[7:].split("?")[0])
+        elif lower.startswith("tel:"):
+            _append_contact(contact_signals, "phones", href[4:].split("?")[0])
+
+    if any(contact_signals[k] for k in contact_signals):
+        out["contact_signals"] = contact_signals
+
     # Empty anchors
     empty_anchors = 0
     for a in soup.find_all("a", href=True):
@@ -363,5 +478,21 @@ def _script_in_head(sc) -> bool:
             "Use <label for=\"id\">, wrap in <label>, or aria-label.",
         )
 
+    pagination: dict[str, str | None] = {"rel_next": None, "rel_prev": None, "amphtml": None}
+    for link in soup.find_all("link", rel=True):
+        rel_raw = link.get("rel") or []
+        rels = {str(r).lower() for r in (rel_raw if isinstance(rel_raw, list) else [rel_raw])}
+        href = link.get("href") or ""
+        if not href:
+            continue
+        abs_href = normalize_link(base_url, href) or href
+        if "next" in rels:
+            pagination["rel_next"] = abs_href
+        if "prev" in rels:
+            pagination["rel_prev"] = abs_href
+        if "amphtml" in rels:
+            pagination["amphtml"] = abs_href
+    out["pagination"] = pagination
+
     out["warnings"] = warnings
     return out
diff --git a/src/website_profiling/commands/pipeline_cmd.py b/src/website_profiling/commands/pipeline_cmd.py
index 4100a3de..6802aa5e 100644
--- a/src/website_profiling/commands/pipeline_cmd.py
+++ b/src/website_profiling/commands/pipeline_cmd.py
@@ -7,6 +7,7 @@
 import pandas as pd
 
 from ..config import get_bool, get_float, get_int, get_list
+from ..progress import emit_phase_done, emit_phase_start
 from .config_resolve import (
     active_property_id_from_cfg,
     cleanup_lighthouse_work_dir,
@@ -96,6 +97,7 @@ def run(cfg: dict, args: argparse.Namespace) -> None:
     if args.command is None and (
         run_crawl or run_lighthouse or run_lighthouse_on_pages or run_report or run_plot
     ):
+        emit_phase_start("config", message="Resolving pipeline configuration")
         steps = []
         if run_crawl:
             steps.append("crawl")
@@ -108,6 +110,7 @@ def run(cfg: dict, args: argparse.Namespace) -> None:
         if run_plot:
             steps.append("plot")
         print(f"Site Audit: {', '.join(steps)}", flush=True)
+        emit_phase_done("config")
 
     if run_crawl:
         _run_crawl(cfg, use_database)
@@ -129,6 +132,7 @@ def _run_crawl(cfg: dict, use_database: bool) -> None:
     from ..crawl.crawler import run_crawler
 
     print("[Crawl] Starting...", flush=True)
+    emit_phase_start("crawl")
     start_url = require_start_url(cfg, for_step="crawl")
     max_pages = get_int(cfg, "max_pages")
     concurrency = get_int(cfg, "concurrency", 8)
@@ -159,6 +163,14 @@ def _run_crawl(cfg: dict, use_database: bool) -> None:
     custom_extraction_regex = (cfg.get("custom_extraction_regex") or "").strip()
     crawl_ignore_raw = (cfg.get("crawl_ignore_params") or "").strip()
     crawl_ignore_params = [p.strip() for p in crawl_ignore_raw.split(",") if p.strip()] or None
+    from ..crawl.discovery import normalize_discovery_mode, parse_crawl_url_list
+
+    discovery_mode = normalize_discovery_mode(cfg.get("crawl_discovery_mode"))
+    crawl_url_list = parse_crawl_url_list(cfg.get("crawl_url_list"), start_url=start_url)
+    from ..crawl.extraction import parse_extractors_config
+
+    custom_extractors = parse_extractors_config(cfg.get("custom_extractors"))
+    enable_axe = get_bool(cfg, "enable_axe", False)
     print("Crawling...")
     run_crawler(
         start_url=start_url,
@@ -191,8 +203,20 @@ def _run_crawl(cfg: dict, use_database: bool) -> None:
         console_max_per_page=console_max_per_page,
         custom_extraction_regex=custom_extraction_regex,
         crawl_ignore_params=crawl_ignore_params,
+        discovery_mode=discovery_mode,
+        crawl_url_list=crawl_url_list or None,
+        crawl_user_agent_preset=(cfg.get("crawl_user_agent_preset") or "default").strip(),
+        crawl_user_agent_custom=(cfg.get("crawl_user_agent_custom") or "").strip(),
+        crawl_auth_username=(cfg.get("crawl_auth_username") or "").strip(),
+        crawl_auth_password=(cfg.get("crawl_auth_password") or "").strip(),
+        crawl_extra_headers=(cfg.get("crawl_extra_headers") or "").strip(),
+        crawl_cookies=(cfg.get("crawl_cookies") or "").strip(),
+        crawl_robots_txt_override=(cfg.get("crawl_robots_txt_override") or "").strip(),
+        custom_extractors=custom_extractors or None,
+        enable_axe=enable_axe,
     )
     print("[Crawl] Done.", flush=True)
+    emit_phase_done("crawl")
     print("Crawl results: PostgreSQL")
 
 
@@ -201,6 +225,7 @@ def _run_lighthouse_on_pages(cfg: dict, lighthouse_max_pages: int) -> None:
     from ..lighthouse.runner import run_lighthouse_on_pages as do_lighthouse_on_pages
 
     print("[Lighthouse on pages] Starting...", flush=True)
+    emit_phase_start("lighthouse", message="Lighthouse on pages")
     with db_session() as conn:
         run_id = get_latest_crawl_run_id(conn)
         df = read_crawl(conn, run_id)
@@ -239,12 +264,14 @@ def _run_lighthouse_on_pages(cfg: dict, lighthouse_max_pages: int) -> None:
         finally:
             cleanup_lighthouse_work_dir(lh_out)
     print("[Lighthouse on pages] Done.", flush=True)
+    emit_phase_done("lighthouse")
 
 
 def _run_single_lighthouse(cfg: dict, use_database: bool) -> None:
     from ..lighthouse.runner import main as lighthouse_main
 
     print("[Lighthouse] Starting...", flush=True)
+    emit_phase_start("lighthouse")
     lh_url = require_lighthouse_url(cfg)
     lh_strategy = (cfg.get("lighthouse_strategy") or "mobile").lower()
     if lh_strategy not in ("mobile", "desktop"):
@@ -268,6 +295,7 @@ def _run_single_lighthouse(cfg: dict, use_database: bool) -> None:
     if exit_code != 0:
         sys.exit(exit_code)
     print("[Lighthouse] Done.", flush=True)
+    emit_phase_done("lighthouse")
 
 
 def _run_report(cfg: dict, use_database: bool) -> None:
@@ -283,6 +311,7 @@ def _run_report(cfg: dict, use_database: bool) -> None:
     security_scan_active = get_bool(cfg, "security_scan_active", False)
     security_max_urls_probe = get_int(cfg, "security_max_urls_probe", 20) or 20
     print("[Report] Starting...", flush=True)
+    emit_phase_start("report")
     out = run_simple_report(
         max_fetch_for_edges=max_fetch,
         concurrency=6,
@@ -300,23 +329,28 @@ def _run_report(cfg: dict, use_database: bool) -> None:
         config=cfg,
     )
     print("[Report] Done.", flush=True)
+    emit_phase_done("report")
     print(f"Report written: {out}")
 
     if should_enrich_keywords_after_report(cfg) and google_db_has_gsc(cfg):
         print("[Keywords] Post-audit keyword research (Search Console data found)...", flush=True)
+        emit_phase_start("keywords")
         from ..integrations.google.keyword_enrich import run_enrichment
 
         try:
             run_enrichment(cfg)
             print("[Keywords] Post-audit keyword research done.", flush=True)
+            emit_phase_done("keywords")
         except Exception as e:
             print(f"Warning: post-audit keyword research failed: {e}", file=sys.stderr)
+            emit_phase_done("keywords", message=f"keywords failed: {e}")
 
 
 def _run_plot(cfg: dict, use_database: bool) -> None:
     from ..tools.plot import run_plot as do_plot
 
     print("[Plot] Starting...", flush=True)
+    emit_phase_start("plot", message="Building charts and link graph")
     render_mode = (cfg.get("crawl_render_mode") or "").strip().lower() or None
     if render_mode is not None and render_mode not in _ALLOWED_RENDER_MODES:
         print(
@@ -331,19 +365,24 @@ def _run_plot(cfg: dict, use_database: bool) -> None:
     if js_extra_wait_ms is None:
         js_extra_wait_ms = 1500
     js_block_resources = get_bool(cfg, "crawl_js_block_resources", True)
-    e = do_plot(
-        same_domain_only=get_bool(cfg, "same_domain_only", True),
-        max_fetch_for_edges=get_int(cfg, "max_fetch_for_edges", 500),
-        concurrency=8,
-        timeout=10,
-        polite_delay=0.15,
-        use_database=use_database,
-        render_mode=render_mode,
-        js_timeout=js_timeout,
-        js_concurrency=js_concurrency,
-        js_wait_until=js_wait_until,
-        js_extra_wait_ms=js_extra_wait_ms,
-        js_block_resources=js_block_resources,
-    )
-    print("[Plot] Done.", flush=True)
-    print(f"Plot data: {e}")
+    try:
+        e = do_plot(
+            same_domain_only=get_bool(cfg, "same_domain_only", True),
+            max_fetch_for_edges=get_int(cfg, "max_fetch_for_edges", 500),
+            concurrency=8,
+            timeout=10,
+            polite_delay=0.15,
+            use_database=use_database,
+            render_mode=render_mode,
+            js_timeout=js_timeout,
+            js_concurrency=js_concurrency,
+            js_wait_until=js_wait_until,
+            js_extra_wait_ms=js_extra_wait_ms,
+            js_block_resources=js_block_resources,
+        )
+        print("[Plot] Done.", flush=True)
+        emit_phase_done("plot", message="Charts and link graph complete")
+        print(f"Plot data: {e}")
+    except Exception:
+        emit_phase_done("plot", message="Charts step failed")
+        raise
diff --git a/src/website_profiling/common.py b/src/website_profiling/common.py
index 82e8cd04..76f451f9 100644
--- a/src/website_profiling/common.py
+++ b/src/website_profiling/common.py
@@ -112,20 +112,61 @@ def normalize_link(
     return out
 
 
-def parse_links(base_url: str, html_text: str) -> tuple[str, set[str]]:
-    """Extract page title and set of absolute links from HTML. Returns (title, links)."""
+def _parse_rel_flags(rel_raw: str) -> tuple[bool, bool, bool]:
+    parts = {p.strip().lower() for p in (rel_raw or "").split() if p.strip()}
+    return ("nofollow" in parts, "sponsored" in parts, "ugc" in parts)
+
+
+def _anchor_text_from_tag(a) -> str:
+    parts: list[str] = []
+    for child in a.children:
+        if getattr(child, "name", None) == "img":
+            parts.append("[image]")
+        elif isinstance(child, str):
+            t = child.strip()
+            if t:
+                parts.append(t)
+    text = " ".join(parts).strip() or a.get_text(separator=" ", strip=True)
+    return (text or "")[:500]
+
+
+def parse_link_edges(base_url: str, html_text: str) -> tuple[str, list[dict]]:
+    """Extract title and rich outbound link records from HTML."""
     soup = BeautifulSoup(html_text, "lxml")
     title_tag = (
         soup.title.string.strip()
         if soup.title and soup.title.string
         else ""
     )
-    links = set()
+    start_netloc = urlparse(base_url).netloc
+    edges: list[dict] = []
     for a in soup.find_all("a", href=True):
         ln = normalize_link(base_url, a["href"])
-        if ln:
-            links.add(ln)
-    return title_tag, links
+        if not ln:
+            continue
+        rel_raw = a.get("rel") or ""
+        if isinstance(rel_raw, list):
+            rel_str = " ".join(str(x) for x in rel_raw)
+        else:
+            rel_str = str(rel_raw)
+        nofollow, sponsored, ugc = _parse_rel_flags(rel_str)
+        link_type = "internal" if urlparse(ln).netloc == start_netloc else "external"
+        edges.append({
+            "to_url": ln.rstrip("/"),
+            "anchor_text": _anchor_text_from_tag(a),
+            "rel": rel_str.strip(),
+            "is_nofollow": nofollow,
+            "is_sponsored": sponsored,
+            "is_ugc": ugc,
+            "link_type": link_type,
+        })
+    return title_tag, edges
+
+
+def parse_links(base_url: str, html_text: str) -> tuple[str, set[str]]:
+    """Extract page title and set of absolute links from HTML. Returns (title, links)."""
+    title, edges = parse_link_edges(base_url, html_text)
+    return title, {e["to_url"] for e in edges}
 
 
 def parse_seo(base_url: str, html_text: str) -> tuple[str, int, str, int, str]:
diff --git a/src/website_profiling/crawl/axe_runner.py b/src/website_profiling/crawl/axe_runner.py
new file mode 100644
index 00000000..28458dbf
--- /dev/null
+++ b/src/website_profiling/crawl/axe_runner.py
@@ -0,0 +1,29 @@
+"""Inject axe-core in Playwright pages when enable_axe is set."""
+from __future__ import annotations
+
+from typing import Any
+
+
+_AXE_CDN = "https://cdnjs.cloudflare.com/ajax/libs/axe-core/4.10.2/axe.min.js"
+
+
+async def run_axe_on_page(page: Any) -> list[dict[str, Any]]:
+    """Run axe-core against the current page; return violation summaries."""
+    try:
+        await page.add_script_tag(url=_AXE_CDN)
+        raw = await page.evaluate(
+            """async () => {
+              if (typeof axe === 'undefined') return [];
+              const results = await axe.run(document, { resultTypes: ['violations'] });
+              return (results.violations || []).slice(0, 20).map(v => ({
+                id: v.id,
+                impact: v.impact,
+                description: v.description,
+                help: v.help,
+                nodes: (v.nodes || []).length
+              }));
+            }"""
+        )
+        return raw if isinstance(raw, list) else []
+    except Exception:
+        return []
diff --git a/src/website_profiling/crawl/crawler.py b/src/website_profiling/crawl/crawler.py
index 19b8c66c..358bc300 100644
--- a/src/website_profiling/crawl/crawler.py
+++ b/src/website_profiling/crawl/crawler.py
@@ -32,7 +32,7 @@ def _url_matches_exclude(url: str, exclude_urls: list[str]) -> bool:
     load_robots,
     normalize_link,
     parse_content_text,
-    parse_links,
+    parse_link_edges,
     parse_resources,
     parse_seo,
     parse_seo_extended,
@@ -40,6 +40,12 @@ def _url_matches_exclude(url: str, exclude_urls: list[str]) -> bool:
     parse_tech_stack,
 )
 from ..analysis.page import analyze_html
+from .discovery import (
+    follow_links_for_mode,
+    normalize_discovery_mode,
+    seed_sitemap_for_mode,
+)
+from .extraction import parse_extractors_config, run_extractors
 from .fetchers import build_fetcher
 from .fetchers.base import FetchResult
 from .fetchers.browser_diagnostics import merge_browser_into_page_analysis
@@ -48,6 +54,19 @@ def _url_matches_exclude(url: str, exclude_urls: list[str]) -> bool:
 from .sitemap import discover_sitemap_urls
 
 DEFAULT_USER_AGENT = "WebsiteProfilingCrawler/1.0"
+MOBILE_USER_AGENT = (
+    "Mozilla/5.0 (iPhone; CPU iPhone OS 17_0 like Mac OS X) "
+    "AppleWebKit/605.1.15 (KHTML, like Gecko) Version/17.0 Mobile/15E148 Safari/604.1"
+)
+
+
+def resolve_crawl_user_agent(preset: str | None, custom: str | None, default: str | None = None) -> str:
+    p = (preset or "default").strip().lower()
+    if p == "mobile":
+        return MOBILE_USER_AGENT
+    if p == "custom" and custom and str(custom).strip():
+        return str(custom).strip()
+    return (default or DEFAULT_USER_AGENT).strip() or DEFAULT_USER_AGENT
 
 
 class Crawler:
@@ -79,9 +98,24 @@ def __init__(
         console_max_per_page: int = 20,
         custom_extraction_regex: str = "",
         crawl_ignore_params: Optional[list[str]] = None,
+        discovery_mode: str = "spider",
+        crawl_url_list: Optional[list[str]] = None,
+        crawl_user_agent_preset: str = "default",
+        crawl_user_agent_custom: str = "",
+        crawl_auth_username: str = "",
+        crawl_auth_password: str = "",
+        crawl_extra_headers: str = "",
+        crawl_cookies: str = "",
+        crawl_robots_txt_override: str = "",
+        custom_extractors: Optional[list[dict]] = None,
+        enable_axe: bool = False,
     ):
         self.start_url = start_url.rstrip("/")
         self.start_netloc = urlparse(self.start_url).netloc
+        self.discovery_mode = normalize_discovery_mode(discovery_mode)
+        self.follow_links = follow_links_for_mode(self.discovery_mode)
+        self.crawl_url_list = [u.rstrip("/") for u in (crawl_url_list or []) if u and str(u).strip()]
+        self.link_edges_accum: list[dict] = []
         self.render_mode = (render_mode or "static").strip().lower()
         self.js_concurrency = max(1, int(js_concurrency))
         effective_concurrency = (
@@ -97,7 +131,9 @@ def __init__(
         self.ignore_robots = ignore_robots
         self.allow_external = allow_external
         self.max_depth = None if max_depth is None else int(max_depth)
-        self.user_agent = user_agent or DEFAULT_USER_AGENT
+        self.user_agent = resolve_crawl_user_agent(
+            crawl_user_agent_preset, crawl_user_agent_custom, user_agent
+        )
         self.polite_delay = max(0.0, float(polite_delay))
         self.store_outlinks = store_outlinks
         self.exclude_urls = list(exclude_urls) if exclude_urls else []
@@ -106,18 +142,37 @@ def __init__(
         self.content_excerpt_max_chars = max(0, int(content_excerpt_max_chars or 0))
         self._wappalyzer_instance = None
         self.custom_extraction_regex = (custom_extraction_regex or "").strip()
+        self.custom_extractors = list(custom_extractors or [])
         self.crawl_ignore_params = list(crawl_ignore_params or [])
 
         self.queue = Queue()
-        if not _url_matches_exclude(self.start_url, self.exclude_urls):
-            self.queue.put(self.start_url)
-        self.depths = {self.start_url: 0}
+        self.depths: dict[str, int] = {}
         self.visited = set()
         self.results = []
         self.lock = threading.Lock()
         self.session = requests.Session()
         self.session.headers.update({"User-Agent": self.user_agent})
-        self.rp = None if self.ignore_robots else load_robots(self.start_url)
+        if crawl_auth_username:
+            self.session.auth = (crawl_auth_username, crawl_auth_password or "")
+        for line in (crawl_extra_headers or "").replace("\r", "").split("\n"):
+            if ":" in line:
+                key, val = line.split(":", 1)
+                k, v = key.strip(), val.strip()
+                if k:
+                    self.session.headers[k] = v
+        if crawl_cookies and str(crawl_cookies).strip():
+            self.session.headers["Cookie"] = str(crawl_cookies).strip()
+        self.rp = None
+        if not self.ignore_robots:
+            override = (crawl_robots_txt_override or "").strip()
+            if override:
+                import io
+                import urllib.robotparser as robotparser
+
+                self.rp = robotparser.RobotFileParser()
+                self.rp.parse(override.splitlines())
+            else:
+                self.rp = load_robots(self.start_url)
         self.fetcher = build_fetcher(
             render_mode="javascript" if self.render_mode == "javascript" else ("auto" if self.render_mode == "auto" else "static"),
             timeout=timeout,
@@ -132,11 +187,33 @@ def __init__(
             js_console_levels=js_console_levels,
             capture_failed_requests=capture_failed_requests,
             console_max_per_page=console_max_per_page,
+            run_axe=enable_axe,
         )
         self._hybrid_fetcher = (
             self.fetcher if isinstance(self.fetcher, HybridFetcher) else None
         )
-        self._seed_sitemap_urls(timeout)
+        self._seed_initial_urls(timeout)
+
+    def _enqueue_seed(self, url: str, depth: int = 0) -> None:
+        u = url.rstrip("/")
+        if _url_matches_exclude(u, self.exclude_urls):
+            return
+        if not self.allow_external and not self.same_domain(u):
+            return
+        if u in self.depths:
+            return
+        self.queue.put(u)
+        self.depths[u] = depth
+
+    def _seed_initial_urls(self, timeout: int) -> None:
+        mode = self.discovery_mode
+        if mode in ("list", "hybrid"):
+            for url in self.crawl_url_list:
+                self._enqueue_seed(url, 0)
+        if mode in ("spider", "hybrid"):
+            self._enqueue_seed(self.start_url, 0)
+        if seed_sitemap_for_mode(mode):
+            self._seed_sitemap_urls(timeout)
 
     def same_domain(self, url):
         return urlparse(url).netloc == self.start_netloc
@@ -159,14 +236,7 @@ def _seed_sitemap_urls(self, timeout: int) -> None:
         except Exception:
             return
         for url in seeds:
-            if _url_matches_exclude(url, self.exclude_urls):
-                continue
-            if not self.allow_external and not self.same_domain(url):
-                continue
-            if url == self.start_url or url in self.depths:
-                continue
-            self.queue.put(url)
-            self.depths[url] = 0
+            self._enqueue_seed(url, 0)
 
     def fetch(self, url) -> FetchResult:
         return self.fetcher.fetch(url)
@@ -233,7 +303,8 @@ def _parse_page_content(
     ) -> dict:
         """Extract title, links, and SEO/content fields from HTML."""
         ext = self._empty_seo(url, headers_dict, redirect_chain_length)
-        title, links = parse_links(url, text)
+        title, link_edge_rows = parse_link_edges(url, text)
+        links = {e["to_url"] for e in link_edge_rows}
         meta_description, meta_description_len, h1_text, h1_count, canonical_url = (
             parse_seo(url, text)
         )
@@ -284,6 +355,7 @@ def _parse_page_content(
         return {
             "title": title,
             "links": links,
+            "link_edges": link_edge_rows,
             "meta_description": meta_description,
             "meta_description_len": meta_description_len,
             "h1_text": h1_text,
@@ -452,7 +524,14 @@ def worker(self, url):
 
                 links = [strip_crawl_query_params(l, self.crawl_ignore_params) for l in links]
 
-            for link in links:
+            link_edge_rows = parsed.get("link_edges") or []
+            for edge in link_edge_rows:
+                link = edge.get("to_url") or ""
+                if self.store_outlinks:
+                    outlink_list.append(link)
+                    self.link_edges_accum.append({"from_url": url, **edge})
+                if not self.follow_links:
+                    continue
                 if _url_matches_exclude(link, self.exclude_urls):
                     continue
                 if not self.allow_external and not self.same_domain(link):
@@ -468,8 +547,6 @@ def worker(self, url):
                     ):
                         self.queue.put(link)
                         self.depths[link] = cur_depth + 1
-                if self.store_outlinks:
-                    outlink_list.append(link)
 
         ext["response_time_ms"] = response_time_ms if response_time_ms is not None else ""
         ext["content_length"] = content_length or 0
@@ -499,6 +576,11 @@ def worker(self, url):
             except re.error:
                 pass
 
+        if self.custom_extractors and text:
+            fields = run_extractors(text, self.custom_extractors)
+            if fields:
+                ext["custom_fields"] = json.dumps(fields)
+
         if self.polite_delay:
             time.sleep(self.polite_delay)
 
@@ -533,6 +615,11 @@ def crawl(
         stream_batch_size: int = 500,
     ):
         start_time = time.time()
+        from ..progress import CrawlProgressTracker, emit_phase_start
+
+        crawl_total = None if self.max_pages == float("inf") else int(self.max_pages)
+        progress_tracker = CrawlProgressTracker(crawl_total, start_time=start_time)
+        emit_phase_start("crawl", message="Crawling pages")
         futures = []
         db_writer: Optional[_CrawlDbWriter] = None
         if stream_crawl_run_id is not None:
@@ -627,6 +714,10 @@ def crawl(
                             if db_writer is not None and res.get("url"):
                                 db_writer.enqueue(res)
                             pbar.update(1)
+                            progress_tracker.maybe_emit(
+                                len(self.results),
+                                str(res.get("url") or "") or None,
+                            )
                         else:
                             remaining.append(f)
                     futures = remaining
@@ -782,15 +873,27 @@ def run_crawler(
     console_max_per_page: int = 20,
     custom_extraction_regex: str = "",
     crawl_ignore_params: Optional[list[str]] = None,
+    discovery_mode: str = "spider",
+    crawl_url_list: Optional[list[str]] = None,
+    crawl_user_agent_preset: str = "default",
+    crawl_user_agent_custom: str = "",
+    crawl_auth_username: str = "",
+    crawl_auth_password: str = "",
+    crawl_extra_headers: str = "",
+    crawl_cookies: str = "",
+    crawl_robots_txt_override: str = "",
+    custom_extractors: Optional[list] = None,
+    enable_axe: bool = False,
 ) -> pd.DataFrame:
     """Run crawler and optionally save to CSV/JSON or PostgreSQL. Returns DataFrame."""
     import sys
     max_p = max_pages if max_pages is not None else 0
     mode_label = (render_mode or "static").strip().lower()
+    disc_label = normalize_discovery_mode(discovery_mode)
     conc_label = js_concurrency if mode_label == "javascript" else concurrency
     print(
         f"  Crawling {start_url} (max_pages={max_p or 'unlimited'}, "
-        f"render_mode={mode_label}, concurrency={conc_label})...",
+        f"discovery={disc_label}, render_mode={mode_label}, concurrency={conc_label})...",
         flush=True,
     )
     crawler = Crawler(
@@ -818,6 +921,17 @@ def run_crawler(
         console_max_per_page=console_max_per_page,
         custom_extraction_regex=custom_extraction_regex,
         crawl_ignore_params=crawl_ignore_params,
+        discovery_mode=disc_label,
+        crawl_url_list=crawl_url_list,
+        crawl_user_agent_preset=crawl_user_agent_preset,
+        crawl_user_agent_custom=crawl_user_agent_custom,
+        crawl_auth_username=crawl_auth_username,
+        crawl_auth_password=crawl_auth_password,
+        crawl_extra_headers=crawl_extra_headers,
+        crawl_cookies=crawl_cookies,
+        crawl_robots_txt_override=crawl_robots_txt_override,
+        custom_extractors=custom_extractors,
+        enable_axe=enable_axe,
     )
     stream_run_id: Optional[int] = None
     if output_db:
@@ -838,7 +952,8 @@ def run_crawler(
                 if historical:
                     restore_historical_data(conn, historical)
                 stream_run_id = create_crawl_run(
-                    conn, start_url, property_id=property_id, render_mode=render_mode
+                    conn, start_url, property_id=property_id, render_mode=render_mode,
+                    discovery_mode=disc_label,
                 )
             print(f"  Streaming crawl results to DB (run_id={stream_run_id})...", flush=True)
 
@@ -846,6 +961,19 @@ def run_crawler(
         show_progress=show_progress,
         stream_crawl_run_id=stream_run_id,
     )
+    if output_db and crawler.link_edges_accum:
+        from ..db import db_session
+        from ..db.crawl_store import write_link_edges
+
+        run_id = stream_run_id
+        if run_id is None:
+            with db_session() as conn:
+                from ..db.crawl_store import get_latest_crawl_run_id
+
+                run_id = get_latest_crawl_run_id(conn)
+        if run_id is not None:
+            with db_session() as conn:
+                write_link_edges(conn, crawler.link_edges_accum, crawl_run_id=run_id)
     if output_db and not df.empty and stream_run_id is None:
         import sys
         print("  Writing crawl results to DB...", flush=True)
@@ -867,9 +995,14 @@ def run_crawler(
             if historical:
                 restore_historical_data(conn, historical)
             run_id = create_crawl_run(
-                conn, start_url, property_id=property_id, render_mode=render_mode
+                conn, start_url, property_id=property_id, render_mode=render_mode,
+                discovery_mode=disc_label,
             )
             write_crawl(conn, df, crawl_run_id=run_id)
+            if crawler.link_edges_accum:
+                from ..db.crawl_store import write_link_edges
+
+                write_link_edges(conn, crawler.link_edges_accum, crawl_run_id=run_id)
         print("  Crawl DB write complete.", flush=True)
     elif output_db and stream_run_id is not None:
         print("  Crawl streamed to DB during fetch.", flush=True)
diff --git a/src/website_profiling/crawl/discovery.py b/src/website_profiling/crawl/discovery.py
new file mode 100644
index 00000000..95efd758
--- /dev/null
+++ b/src/website_profiling/crawl/discovery.py
@@ -0,0 +1,40 @@
+"""Crawl discovery mode helpers (spider, list, sitemap, hybrid)."""
+from __future__ import annotations
+
+from ..common import normalize_link
+
+DISCOVERY_MODES = frozenset({"spider", "list", "sitemap", "hybrid"})
+
+
+def normalize_discovery_mode(mode: str | None) -> str:
+    m = (mode or "spider").strip().lower()
+    return m if m in DISCOVERY_MODES else "spider"
+
+
+def parse_crawl_url_list(raw: str | None, *, start_url: str = "") -> list[str]:
+    """Parse newline- or comma-separated URL list; dedupe preserving order."""
+    if not raw or not str(raw).strip():
+        return []
+    seen: set[str] = set()
+    out: list[str] = []
+    for line in str(raw).replace("\r", "").split("\n"):
+        for part in line.split(","):
+            u = part.strip().rstrip("/")
+            if not u:
+                continue
+            if start_url and not u.startswith(("http://", "https://")):
+                normalized = normalize_link(start_url, u)
+                if normalized:
+                    u = normalized.rstrip("/")
+            if u and u not in seen:
+                seen.add(u)
+                out.append(u)
+    return out
+
+
+def follow_links_for_mode(mode: str) -> bool:
+    return normalize_discovery_mode(mode) in ("spider", "hybrid")
+
+
+def seed_sitemap_for_mode(mode: str) -> bool:
+    return normalize_discovery_mode(mode) in ("spider", "sitemap", "hybrid")
diff --git a/src/website_profiling/crawl/extraction.py b/src/website_profiling/crawl/extraction.py
new file mode 100644
index 00000000..ccb9180b
--- /dev/null
+++ b/src/website_profiling/crawl/extraction.py
@@ -0,0 +1,65 @@
+"""Run configured HTML extractors (regex, css, xpath)."""
+from __future__ import annotations
+
+import json
+import re
+from typing import Any
+
+from bs4 import BeautifulSoup
+
+
+def parse_extractors_config(raw: str | None) -> list[dict[str, Any]]:
+    if not raw or not str(raw).strip():
+        return []
+    try:
+        data = json.loads(str(raw))
+        return [x for x in data if isinstance(x, dict)] if isinstance(data, list) else []
+    except json.JSONDecodeError:
+        return []
+
+
+def run_extractors(html: str, extractors: list[dict[str, Any]]) -> dict[str, str]:
+    if not html or not extractors:
+        return {}
+    soup = BeautifulSoup(html, "lxml")
+    out: dict[str, str] = {}
+    for spec in extractors:
+        name = str(spec.get("name") or "").strip()
+        if not name:
+            continue
+        kind = str(spec.get("type") or "regex").strip().lower()
+        try:
+            if kind == "css":
+                sel = str(spec.get("selector") or "").strip()
+                if not sel:
+                    continue
+                el = soup.select_one(sel)
+                if el is None:
+                    continue
+                attr = str(spec.get("attr") or "").strip()
+                val = el.get(attr) if attr else el.get_text(strip=True)
+                if val:
+                    out[name] = str(val)[:500]
+            elif kind == "xpath":
+                from lxml import html as lxml_html
+
+                expr = str(spec.get("expr") or spec.get("xpath") or "").strip()
+                if not expr:
+                    continue
+                tree = lxml_html.fromstring(html)
+                hits = tree.xpath(expr)
+                if hits:
+                    first = hits[0]
+                    val = first if isinstance(first, str) else getattr(first, "text_content", lambda: "")()
+                    if val:
+                        out[name] = str(val).strip()[:500]
+            else:
+                pattern = str(spec.get("pattern") or spec.get("regex") or "").strip()
+                if not pattern:
+                    continue
+                match = re.search(pattern, html)
+                if match:
+                    out[name] = (match.group(1) if match.lastindex else match.group(0))[:500]
+        except Exception:
+            continue
+    return out
diff --git a/src/website_profiling/crawl/fetchers/browser.py b/src/website_profiling/crawl/fetchers/browser.py
index c91a764f..61c7e3c0 100644
--- a/src/website_profiling/crawl/fetchers/browser.py
+++ b/src/website_profiling/crawl/fetchers/browser.py
@@ -139,9 +139,14 @@ def __init__(
         console_levels: frozenset[str] | None = None,
         capture_failed_requests: bool = False,
         console_max_per_page: int = 20,
+        run_axe: bool = False,
+        extra_http_headers: Optional[dict[str, str]] = None,
+        http_credentials: Optional[dict[str, str]] = None,
     ) -> None:
         self.timeout = max(1, int(timeout))
         self.user_agent = user_agent
+        self.extra_http_headers = dict(extra_http_headers or {})
+        self.http_credentials = http_credentials
         self.js_concurrency = max(1, int(js_concurrency))
         self.wait_until = wait_until if wait_until in ("domcontentloaded", "load", "commit") else "domcontentloaded"
         self.extra_wait_ms = max(0, int(extra_wait_ms))
@@ -150,6 +155,7 @@ def __init__(
         self.console_levels = console_levels or frozenset({"error", "warning"})
         self.capture_failed_requests = bool(capture_failed_requests)
         self.console_max_per_page = max(1, int(console_max_per_page))
+        self.run_axe = bool(run_axe)
 
         self._loop: Optional[asyncio.AbstractEventLoop] = None
         self._thread: Optional[threading.Thread] = None
@@ -199,7 +205,12 @@ async def _async_main(self) -> None:
             launch_kwargs["executable_path"] = chrome_path
 
         browser = await playwright.chromium.launch(**launch_kwargs)
-        context = await browser.new_context(user_agent=self.user_agent)
+        context_kwargs: dict[str, Any] = {"user_agent": self.user_agent}
+        if self.extra_http_headers:
+            context_kwargs["extra_http_headers"] = self.extra_http_headers
+        if self.http_credentials:
+            context_kwargs["http_credentials"] = self.http_credentials
+        context = await browser.new_context(**context_kwargs)
         semaphore = asyncio.Semaphore(self.js_concurrency)
         pages: list[Any] = []
         for _ in range(self.js_concurrency):
@@ -339,6 +350,14 @@ async def _fetch_page(self, page: Any, url: str) -> FetchResult:
                 content_length = len(text.encode("utf-8")) if text else 0
             except Exception:
                 text = None
+            if self.run_axe and text:
+                from ..axe_runner import run_axe_on_page
+
+                axe_violations = await run_axe_on_page(page)
+                if axe_violations:
+                    if browser_diagnostics is None:
+                        browser_diagnostics = finalize_browser_diagnostics([], [], [])
+                    browser_diagnostics["axe_violations"] = axe_violations
 
         return FetchResult(
             status=status,
diff --git a/src/website_profiling/crawl/fetchers/browser_diagnostics.py b/src/website_profiling/crawl/fetchers/browser_diagnostics.py
index e0d8671e..3f551304 100644
--- a/src/website_profiling/crawl/fetchers/browser_diagnostics.py
+++ b/src/website_profiling/crawl/fetchers/browser_diagnostics.py
@@ -55,6 +55,9 @@ def merge_browser_into_page_analysis(
         except json.JSONDecodeError:
             pa = {}
     pa["browser"] = browser_diagnostics
+    axe = browser_diagnostics.get("axe_violations")
+    if isinstance(axe, list) and axe:
+        pa["axe_violations"] = axe
     return json.dumps(pa)
 
 
diff --git a/src/website_profiling/crawl/fetchers/factory.py b/src/website_profiling/crawl/fetchers/factory.py
index d4086dd2..5f0bb8d5 100644
--- a/src/website_profiling/crawl/fetchers/factory.py
+++ b/src/website_profiling/crawl/fetchers/factory.py
@@ -16,6 +16,24 @@
 RenderMode = Literal["static", "javascript", "auto"]
 
 
+def _browser_auth_from_session(
+    session: Optional[requests.Session],
+) -> tuple[dict[str, str], Optional[dict[str, str]]]:
+    """Map requests session auth/headers onto Playwright browser context options."""
+    if session is None:
+        return {}, None
+    headers: dict[str, str] = {}
+    for key, value in session.headers.items():
+        if str(key).lower() == "user-agent":
+            continue
+        headers[str(key)] = str(value)
+    credentials: Optional[dict[str, str]] = None
+    auth = getattr(session, "auth", None)
+    if auth and auth[0]:
+        credentials = {"username": str(auth[0]), "password": str(auth[1] or "")}
+    return headers, credentials
+
+
 def validate_browser_available() -> None:
     """Raise RuntimeError if JS crawl prerequisites are missing."""
     status = ensure_browser_deps()
@@ -35,6 +53,9 @@ def _browser_factory(
     console_levels: frozenset[str] | None = None,
     capture_failed_requests: bool = False,
     console_max_per_page: int = 20,
+    run_axe: bool = False,
+    extra_http_headers: Optional[dict[str, str]] = None,
+    http_credentials: Optional[dict[str, str]] = None,
 ) -> Callable[[], PageFetcher]:
     def _make() -> PageFetcher:
         return BrowserFetcher(
@@ -48,6 +69,9 @@ def _make() -> PageFetcher:
             console_levels=console_levels,
             capture_failed_requests=capture_failed_requests,
             console_max_per_page=console_max_per_page,
+            run_axe=run_axe,
+            extra_http_headers=extra_http_headers or {},
+            http_credentials=http_credentials,
         )
 
     return _make
@@ -68,9 +92,11 @@ def build_fetcher(
     js_console_levels: str = "error,warning",
     capture_failed_requests: bool = False,
     console_max_per_page: int = 20,
+    run_axe: bool = False,
 ) -> PageFetcher:
     mode = (render_mode or "static").strip().lower()
     levels = parse_console_levels(js_console_levels)
+    extra_http_headers, http_credentials = _browser_auth_from_session(session)
     browser_kwargs = dict(
         js_timeout=js_timeout,
         user_agent=user_agent,
@@ -82,6 +108,9 @@ def build_fetcher(
         console_levels=levels,
         capture_failed_requests=capture_failed_requests,
         console_max_per_page=console_max_per_page,
+        run_axe=run_axe,
+        extra_http_headers=extra_http_headers,
+        http_credentials=http_credentials,
     )
     if mode == "javascript":
         validate_browser_available()
diff --git a/src/website_profiling/db/crawl_store.py b/src/website_profiling/db/crawl_store.py
index 81438225..a4122d17 100644
--- a/src/website_profiling/db/crawl_store.py
+++ b/src/website_profiling/db/crawl_store.py
@@ -30,21 +30,40 @@ def create_crawl_run(
     start_url: Optional[str] = None,
     property_id: Optional[int] = None,
     render_mode: Optional[str] = None,
+    discovery_mode: Optional[str] = None,
 ) -> int:
     mode = (render_mode or "static").strip().lower()
-    try:
-        cur = conn.execute(
+    disc = (discovery_mode or "spider").strip().lower()
+    statements = [
+        (
+            "INSERT INTO crawl_runs (created_at, start_url, property_id, render_mode, discovery_mode) VALUES (%s, %s, %s, %s, %s) RETURNING id",
+            (_now_iso(), start_url, property_id, mode, disc),
+        ),
+        (
             "INSERT INTO crawl_runs (created_at, start_url, property_id, render_mode) VALUES (%s, %s, %s, %s) RETURNING id",
             (_now_iso(), start_url, property_id, mode),
-        )
-    except Exception:
-        cur = conn.execute(
+        ),
+        (
             "INSERT INTO crawl_runs (created_at, start_url, property_id) VALUES (%s, %s, %s) RETURNING id",
             (_now_iso(), start_url, property_id),
-        )
-    row = cur.fetchone()
-    conn.commit()
-    return int(row["id"])
+        ),
+    ]
+    last_err: Exception | None = None
+    for sql, params in statements:
+        try:
+            cur = conn.execute(sql, params)
+            row = cur.fetchone()
+            conn.commit()
+            return int(row["id"])
+        except Exception as exc:
+            last_err = exc
+            try:
+                conn.rollback()
+            except Exception:
+                pass
+    if last_err is not None:
+        raise last_err
+    raise RuntimeError("create_crawl_run failed")  # pragma: no cover
 
 
 def get_latest_crawl_run_id(conn: Connection) -> Optional[int]:
@@ -316,6 +335,67 @@ def write_edges(conn: Connection, edges: list[tuple[str, str]], crawl_run_id: Op
     conn.commit()
 
 
+def write_link_edges(
+    conn: Connection,
+    edges: list[dict],
+    crawl_run_id: Optional[int] = None,
+) -> None:
+    if crawl_run_id is None:
+        crawl_run_id = get_latest_crawl_run_id(conn)
+    if crawl_run_id is None or not edges:
+        return
+    conn.execute("DELETE FROM link_edges WHERE crawl_run_id = %s", (crawl_run_id,))
+    rows = []
+    for e in edges:
+        from_u = str(e.get("from_url") or "").rstrip("/")
+        to_u = str(e.get("to_url") or "").rstrip("/")
+        if not from_u or not to_u:
+            continue
+        rows.append((
+            crawl_run_id,
+            from_u,
+            to_u,
+            str(e.get("anchor_text") or "")[:500],
+            str(e.get("rel") or "")[:200],
+            bool(e.get("is_nofollow")),
+            bool(e.get("is_sponsored")),
+            bool(e.get("is_ugc")),
+            str(e.get("link_type") or "internal"),
+        ))
+    if rows:
+        _executemany(
+            conn,
+            """INSERT INTO link_edges (
+                crawl_run_id, from_url, to_url, anchor_text, rel,
+                is_nofollow, is_sponsored, is_ugc, link_type
+            ) VALUES (%s, %s, %s, %s, %s, %s, %s, %s, %s)
+            ON CONFLICT DO NOTHING""",
+            rows,
+        )
+    conn.commit()
+
+
+def read_link_edges(
+    conn: Connection,
+    run_id: Optional[int] = None,
+    *,
+    limit: int = 5000,
+) -> list[dict[str, Any]]:
+    if run_id is None:
+        run_id = get_latest_crawl_run_id(conn)
+    if run_id is None:
+        return []
+    try:
+        cur = conn.execute(
+            """SELECT from_url, to_url, anchor_text, rel, is_nofollow, is_sponsored, is_ugc, link_type
+               FROM link_edges WHERE crawl_run_id = %s LIMIT %s""",
+            (run_id, max(1, int(limit))),
+        )
+        return [dict(row) for row in cur.fetchall()]
+    except Exception:
+        return []
+
+
 def read_edges(conn: Connection, run_id: Optional[int] = None) -> list[tuple[str, str]]:
     try:
         if run_id is None:
diff --git a/src/website_profiling/integrations/google/rich_results.py b/src/website_profiling/integrations/google/rich_results.py
new file mode 100644
index 00000000..28e559e3
--- /dev/null
+++ b/src/website_profiling/integrations/google/rich_results.py
@@ -0,0 +1,234 @@
+"""Optional Google Rich Results validation via GSC, API, or crawl heuristics."""
+from __future__ import annotations
+
+from typing import Any, Optional
+
+import requests
+
+_JSON_LD_CODES = frozenset({"json_ld_parse", "json_ld_missing_type"})
+_RICH_RESULTS_API = "https://searchconsole.googleapis.com/v1/urlTestingTools/richResults:run"
+_API_MAX_CALLS = 10
+
+
+def _local_row(url: str, link: dict[str, Any] | None) -> dict[str, Any]:
+    if not link:
+        return {
+            "url": url,
+            "status": "skipped",
+            "provenance": "Estimated",
+            "source": "crawl",
+            "message": "No crawl data for URL.",
+        }
+    pa = link.get("page_analysis")
+    if not isinstance(pa, dict):
+        pa = {}
+    warnings = pa.get("warnings") if isinstance(pa.get("warnings"), list) else []
+    schema_warnings = [
+        w for w in warnings
+        if isinstance(w, dict) and str(w.get("code") or "") in _JSON_LD_CODES
+    ]
+    has_schema = bool(link.get("has_schema"))
+    schema_types = pa.get("json_ld_types") or pa.get("schema_types") or []
+    if isinstance(schema_types, str):
+        schema_types = [schema_types]
+
+    if schema_warnings:
+        first = schema_warnings[0]
+        return {
+            "url": url,
+            "status": "warning",
+            "provenance": "Crawl analysis",
+            "source": "crawl",
+            "message": str(first.get("message") or "JSON-LD issue detected during crawl."),
+            "issues": [str(w.get("message") or w.get("code") or "warning") for w in schema_warnings[:5]],
+        }
+    if has_schema and not schema_types:
+        return {
+            "url": url,
+            "status": "warning",
+            "provenance": "Crawl analysis",
+            "source": "crawl",
+            "message": "Structured data detected but JSON-LD @type could not be parsed.",
+        }
+    if has_schema:
+        types = ", ".join(str(t) for t in schema_types[:5])
+        return {
+            "url": url,
+            "status": "pass",
+            "provenance": "Crawl analysis",
+            "source": "crawl",
+            "message": f"JSON-LD present ({types or 'types unknown'}).",
+            "schema_types": list(schema_types)[:10],
+        }
+    return {
+        "url": url,
+        "status": "info",
+        "provenance": "Crawl analysis",
+        "source": "crawl",
+        "message": "No structured data detected on crawled HTML.",
+    }
+
+
+def _inspect_via_gsc(creds: Any, site_url: str, url: str) -> dict[str, Any]:
+    from .gsc import _build_service, _call_with_retry, resolve_gsc_site_url, list_gsc_sites
+
+    sites = list_gsc_sites(creds)
+    resolved, err = resolve_gsc_site_url(site_url, sites)
+    if not resolved:
+        return {
+            "url": url,
+            "status": "error",
+            "provenance": "Google Search Console",
+            "source": "gsc",
+            "message": err or "GSC site URL not accessible.",
+        }
+    service = _build_service(creds)
+    body = {"inspectionUrl": url, "siteUrl": resolved}
+    resp = _call_with_retry(
+        lambda: service.urlInspection().index().inspect(body=body).execute()
+    )
+    inspection = resp.get("inspectionResult") or {}
+    rich = inspection.get("richResultsResult") or {}
+    verdict = str(rich.get("verdict") or "UNKNOWN")
+    detected = rich.get("detectedItems") or []
+    types: list[str] = []
+    for item in detected:
+        if isinstance(item, dict):
+            rt = item.get("richResultType")
+            if rt:
+                types.append(str(rt))
+    status = "pass"
+    if verdict in ("FAIL", "ERROR"):
+        status = "fail"
+    elif verdict in ("PARTIAL", "WARN", "WARNING"):
+        status = "warning"
+    elif verdict in ("NEUTRAL", "UNKNOWN", "NOT_APPLICABLE"):
+        status = "info"
+    message = f"Rich Results verdict: {verdict}"
+    if types:
+        message += f" ({', '.join(types[:5])})"
+    row: dict[str, Any] = {
+        "url": url,
+        "status": status,
+        "provenance": "Google Search Console",
+        "source": "gsc",
+        "message": message,
+        "verdict": verdict,
+    }
+    if types:
+        row["schema_types"] = types[:10]
+    issues = rich.get("issues") or []
+    if issues:
+        row["issues"] = [
+            str(i.get("issueMessage") or i.get("severity") or i)
+            for i in issues[:5]
+            if isinstance(i, dict) or i
+        ]
+    return row
+
+
+def _inspect_via_rich_results_api(api_key: str, url: str) -> dict[str, Any]:
+    """Call Google Rich Results Test API (API key)."""
+    resp = requests.post(
+        _RICH_RESULTS_API,
+        params={"key": api_key.strip()},
+        json={"inspectionUrl": url},
+        timeout=25,
+    )
+    if resp.status_code >= 400:
+        raise RuntimeError(f"Rich Results API HTTP {resp.status_code}: {resp.text[:200]}")
+    data = resp.json()
+    outcome = data.get("richResultsOutcome") or data.get("testStatus") or {}
+    verdict = str(outcome.get("verdict") or outcome.get("status") or "UNKNOWN")
+    status = "pass"
+    if verdict.upper() in ("FAIL", "ERROR", "FAILED"):
+        status = "fail"
+    elif verdict.upper() in ("PARTIAL", "WARN", "WARNING"):
+        status = "warning"
+    elif verdict.upper() in ("NEUTRAL", "UNKNOWN", "NOT_APPLICABLE", "INCOMPLETE"):
+        status = "info"
+    types: list[str] = []
+    for item in outcome.get("detectedItems") or data.get("detectedItems") or []:
+        if isinstance(item, dict):
+            rt = item.get("richResultType") or item.get("type")
+            if rt:
+                types.append(str(rt))
+    message = f"Rich Results API verdict: {verdict}"
+    if types:
+        message += f" ({', '.join(types[:5])})"
+    row: dict[str, Any] = {
+        "url": url,
+        "status": status,
+        "provenance": "Google Rich Results API",
+        "source": "api",
+        "message": message,
+        "verdict": verdict,
+    }
+    if types:
+        row["schema_types"] = types[:10]
+    return row
+
+
+def summarize_rich_results(rows: list[dict[str, Any]]) -> dict[str, int]:
+    meta = {"checked": len(rows), "gsc_count": 0, "api_count": 0, "heuristic_count": 0}
+    for row in rows:
+        src = str(row.get("source") or "")
+        prov = str(row.get("provenance") or "")
+        if src == "gsc" or "Search Console" in prov:
+            meta["gsc_count"] += 1
+        elif src == "api" or "Rich Results API" in prov:
+            meta["api_count"] += 1
+        else:
+            meta["heuristic_count"] += 1
+    return meta
+
+
+def validate_urls(
+    urls: list[str],
+    api_key: str | None = None,
+    *,
+    creds: Any = None,
+    site_url: str | None = None,
+    links_by_url: Optional[dict[str, dict[str, Any]]] = None,
+    api_max_calls: int = _API_MAX_CALLS,
+) -> list[dict[str, Any]]:
+    """Validate Rich Results for sample URLs.
+
+    Fallback chain: GSC URL Inspection → Rich Results Test API → crawl heuristics.
+    """
+    links_by_url = links_by_url or {}
+    rows: list[dict[str, Any]] = []
+    use_gsc = bool(creds and (site_url or "").strip())
+    api_key = (api_key or "").strip()
+    api_calls = 0
+    for url in urls[:50]:
+        u = str(url or "").strip()
+        if not u:
+            continue
+        if use_gsc:
+            try:
+                rows.append(_inspect_via_gsc(creds, str(site_url).strip(), u))
+                continue
+            except Exception as exc:
+                link = links_by_url.get(u) or links_by_url.get(u.rstrip("/"))
+                if api_key and api_calls < api_max_calls:
+                    try:
+                        rows.append(_inspect_via_rich_results_api(api_key, u))
+                        api_calls += 1
+                        continue
+                    except Exception:
+                        pass
+                local = _local_row(u, link if isinstance(link, dict) else None)
+                local["status"] = "error"
+                local["message"] = f"GSC inspection failed: {exc}"
+                rows.append(local)
+                continue
+        if api_key and api_calls < api_max_calls:
+            try:
+                rows.append(_inspect_via_rich_results_api(api_key, u))
+                api_calls += 1
+                continue
+            except Exception:
+                pass
+        rows.append(_local_row(u, links_by_url.get(u) or links_by_url.get(u.rstrip("/"))))
+    return rows
diff --git a/src/website_profiling/integrations/keywords/competitor_csv.py b/src/website_profiling/integrations/keywords/competitor_csv.py
new file mode 100644
index 00000000..9858d582
--- /dev/null
+++ b/src/website_profiling/integrations/keywords/competitor_csv.py
@@ -0,0 +1,32 @@
+"""Parse competitor keyword exports (Ahrefs / Semrush CSV)."""
+from __future__ import annotations
+
+import csv
+import io
+from typing import Any
+
+
+def parse_competitor_keyword_csv(csv_text: str, *, competitor: str = "") -> list[dict[str, Any]]:
+    reader = csv.DictReader(io.StringIO(csv_text or ""))
+    if not reader.fieldnames:
+        return []
+    rows: list[dict[str, Any]] = []
+    for raw in reader:
+        kw = (
+            raw.get("Keyword")
+            or raw.get("keyword")
+            or raw.get("Query")
+            or raw.get("query")
+            or ""
+        ).strip()
+        if not kw:
+            continue
+        rows.append({
+            "keyword": kw,
+            "competitor": competitor,
+            "volume": raw.get("Volume") or raw.get("volume") or raw.get("Search Volume"),
+            "position": raw.get("Position") or raw.get("position") or raw.get("Current position"),
+            "url": raw.get("URL") or raw.get("url") or raw.get("Landing page"),
+            "source": "competitor_csv",
+        })
+    return rows
diff --git a/src/website_profiling/lighthouse/runner.py b/src/website_profiling/lighthouse/runner.py
index defa67e8..9b8517f4 100644
--- a/src/website_profiling/lighthouse/runner.py
+++ b/src/website_profiling/lighthouse/runner.py
@@ -503,14 +503,24 @@ def _audit_one(url: str) -> None:
                 except OSError:
                     pass
 
+    from ..progress import emit_progress
+
     if workers == 1:
         for idx, url in enumerate(urls):
             try:
                 print(f"[Lighthouse on pages] {idx + 1}/{total}: {url}", flush=True)
+                emit_progress(
+                    "lighthouse",
+                    "audit",
+                    current=idx + 1,
+                    total=total,
+                    url=url,
+                )
                 _audit_one(url)
             except Exception as e:
                 print(f"  Skipped (error): {e}", file=sys.stderr, flush=True)
     else:
+        completed = 0
         with ThreadPoolExecutor(max_workers=workers) as pool:
             futures = {pool.submit(_audit_one, url): url for url in urls}
             for future in as_completed(futures):
@@ -519,6 +529,14 @@ def _audit_one(url: str) -> None:
                     future.result()
                 except Exception as e:
                     print(f"  Skipped {url} (error): {e}", file=sys.stderr, flush=True)
+                completed += 1
+                emit_progress(
+                    "lighthouse",
+                    "audit",
+                    current=completed,
+                    total=total,
+                    url=url,
+                )
 
     print(f"[Lighthouse on pages] Done. Wrote {total} URL(s) to DB.", flush=True)
 
diff --git a/src/website_profiling/llm/fix_suggestions.py b/src/website_profiling/llm/fix_suggestions.py
new file mode 100644
index 00000000..004261ef
--- /dev/null
+++ b/src/website_profiling/llm/fix_suggestions.py
@@ -0,0 +1,94 @@
+"""Unified on-demand LLM fix suggestions across audit surfaces."""
+from __future__ import annotations
+
+import hashlib
+import json
+from typing import Any
+
+from ..llm_config import llm_is_enabled
+from .base import get_llm_client, parse_json_response
+from .enrich import _read_cache, _write_cache
+from .prompts import FIX_SUGGESTION_PROMPTS, PROMPT_VERSION
+
+VALID_SOURCES = frozenset(FIX_SUGGESTION_PROMPTS.keys())
+DEFAULT_FIX = {"fix": "Review the issue on the affected URL and apply standard remediation.", "effort": "medium"}
+
+
+def _fix_suggestion_enabled(cfg: dict[str, str]) -> bool:
+    v = str(cfg.get("llm_enable_issue_fixes", "true")).lower()
+    return v in ("true", "1", "yes")
+
+
+def _normalize_source(raw: Any) -> str:
+    source = str(raw or "issue").strip().lower()
+    return source if source in VALID_SOURCES else "issue"
+
+
+def _cache_key(model: str, source: str, payload: dict[str, Any]) -> str:
+    body = json.dumps(payload, sort_keys=True, default=str)
+    digest = hashlib.sha256(f"fix_suggestion:{PROMPT_VERSION}:{model}:{source}:{body}".encode()).hexdigest()
+    return digest
+
+
+def _build_user_payload(payload: dict[str, Any]) -> dict[str, Any]:
+    source = _normalize_source(payload.get("source"))
+    out: dict[str, Any] = {
+        "source": source,
+        "message": str(payload.get("message") or "").strip(),
+    }
+    if payload.get("url"):
+        out["url"] = payload.get("url")
+    context = payload.get("context")
+    if isinstance(context, dict) and context:
+        out["context"] = context
+    # Legacy issue-fix fields folded into context for source=issue
+    if source == "issue":
+        legacy: dict[str, Any] = {}
+        for key in ("priority", "category", "type", "finding_type", "recommendation", "existing_recommendation"):
+            if payload.get(key) is not None:
+                legacy[key] = payload.get(key)
+        if legacy:
+            out.setdefault("context", {}).update(legacy)
+    return out
+
+
+def generate_fix_suggestion(
+    payload: dict[str, Any],
+    *,
+    cfg: dict[str, str] | None = None,
+    refresh: bool = False,
+) -> dict[str, Any]:
+    from ..llm_config import load_llm_config_from_db
+
+    cfg = cfg or load_llm_config_from_db()
+    if not llm_is_enabled(cfg):
+        return {"ok": False, "error": "AI insights are disabled."}
+    if not _fix_suggestion_enabled(cfg):
+        return {"ok": False, "error": "Issue fix suggestions are disabled in AI task settings."}
+
+    user_payload = _build_user_payload(payload)
+    message = user_payload.get("message") or ""
+    if not message:
+        return {"ok": False, "error": "message required."}
+
+    source = str(user_payload["source"])
+    model = (cfg.get("llm_model") or cfg.get("llm_provider") or "unknown").strip()
+    cache_key = _cache_key(model, source, user_payload)
+
+    if not refresh:
+        cached = _read_cache(cache_key)
+        if cached:
+            return {"ok": True, "cached": True, "fix": cached, "provenance": "AI insights"}
+
+    system = FIX_SUGGESTION_PROMPTS[source]
+    try:
+        client = get_llm_client(cfg)
+        user = json.dumps(user_payload, indent=2, default=str)[:8000]
+        raw = client.complete_json(system, user)
+        fix = raw if isinstance(raw, dict) and raw else parse_json_response(str(raw))
+        if not fix or not str(fix.get("fix") or "").strip():
+            fix = dict(DEFAULT_FIX)
+        _write_cache(cache_key, fix)
+        return {"ok": True, "cached": False, "fix": fix, "provenance": "AI insights"}
+    except Exception as e:
+        return {"ok": False, "error": str(e)}
diff --git a/src/website_profiling/llm/issue_fixes.py b/src/website_profiling/llm/issue_fixes.py
index e50d3fce..c592aa62 100644
--- a/src/website_profiling/llm/issue_fixes.py
+++ b/src/website_profiling/llm/issue_fixes.py
@@ -1,19 +1,10 @@
 """LLM-generated fix suggestions for audit issues."""
 from __future__ import annotations
 
-import hashlib
-import json
 from typing import Any
 
 from ..llm_config import llm_is_enabled
-from .base import get_llm_client, parse_json_response
-from .enrich import _read_cache, _write_cache
-from .prompts import ISSUE_FIX_SYSTEM, PROMPT_VERSION
-
-
-def _issue_fix_enabled(cfg: dict[str, str]) -> bool:
-    v = str(cfg.get("llm_enable_issue_fixes", "true")).lower()
-    return v in ("true", "1", "yes")
+from .fix_suggestions import _fix_suggestion_enabled, generate_fix_suggestion
 
 
 def generate_issue_fix_suggestion(
@@ -22,47 +13,21 @@ def generate_issue_fix_suggestion(
     cfg: dict[str, str] | None = None,
     refresh: bool = False,
 ) -> dict[str, Any]:
-    from ..llm_config import load_llm_config_from_db
-
-    cfg = cfg or load_llm_config_from_db()
-    if not llm_is_enabled(cfg):
-        return {"ok": False, "error": "AI insights are disabled."}
-    if not _issue_fix_enabled(cfg):
-        return {"ok": False, "error": "Issue fix suggestions are disabled in AI task settings."}
-
     message = str(issue.get("message") or "").strip()
     if not message:
         return {"ok": False, "error": "Issue message required."}
 
-    payload = {
+    payload: dict[str, Any] = {
+        "source": "issue",
         "message": message,
         "url": issue.get("url"),
+        "refresh": refresh,
         "priority": issue.get("priority"),
         "category": issue.get("category"),
-        "existing_recommendation": issue.get("recommendation"),
+        "recommendation": issue.get("recommendation"),
         "type": issue.get("type") or issue.get("finding_type"),
     }
-    model = (cfg.get("llm_model") or cfg.get("llm_provider") or "unknown").strip()
-    cache_key = hashlib.sha256(
-        f"issue_fix:{PROMPT_VERSION}:{model}:{json.dumps(payload, sort_keys=True)}".encode()
-    ).hexdigest()
-
-    if not refresh:
-        cached = _read_cache(cache_key)
-        if cached:
-            return {"ok": True, "cached": True, "fix": cached, "provenance": "AI insights"}
-
-    try:
-        client = get_llm_client(cfg)
-        user = json.dumps(payload, indent=2, default=str)[:8000]
-        raw = client.complete_json(ISSUE_FIX_SYSTEM, user)
-        fix = raw if isinstance(raw, dict) and raw else parse_json_response(str(raw))
-        if not fix:
-            fix = {"fix": "Review the issue on the affected URL and apply standard SEO remediation."}
-        _write_cache(cache_key, fix)
-        return {"ok": True, "cached": False, "fix": fix, "provenance": "AI insights"}
-    except Exception as e:
-        return {"ok": False, "error": str(e)}
+    return generate_fix_suggestion(payload, cfg=cfg, refresh=refresh)
 
 
 def enrich_top_issues_with_llm(
@@ -75,7 +40,7 @@ def enrich_top_issues_with_llm(
     """Attach llm_recommendation to top traffic-weighted issues in-place."""
     from .audit_summary import rank_issues_by_traffic
 
-    if not cfg or not llm_is_enabled(cfg) or not _issue_fix_enabled(cfg):
+    if not cfg or not llm_is_enabled(cfg) or not _fix_suggestion_enabled(cfg):
         return
 
     ranked = rank_issues_by_traffic(categories, gsc_pages)[:limit]
@@ -84,7 +49,6 @@ def enrich_top_issues_with_llm(
 
     by_key: dict[tuple[str, str], dict[str, Any]] = {}
     for cat in categories or []:
-        cat_name = str(cat.get("name") or cat.get("id") or "")
         for issue in cat.get("issues") or []:
             if not isinstance(issue, dict):
                 continue
diff --git a/src/website_profiling/llm/prompts.py b/src/website_profiling/llm/prompts.py
index fe5f5744..7ab6e720 100644
--- a/src/website_profiling/llm/prompts.py
+++ b/src/website_profiling/llm/prompts.py
@@ -1,7 +1,7 @@
 """Versioned prompts for LLM enrichment tasks."""
 from __future__ import annotations
 
-PROMPT_VERSION = "v1"
+PROMPT_VERSION = "v2"
 
 NER_SYSTEM = """You extract named entities from web page text for SEO analysis.
 Return JSON: {"pages": [{"url": "...", "entity_count": N, "top_entity_labels": [["ORG", 2], ["PERSON", 1]]}]}
@@ -35,6 +35,44 @@
 Use ONLY the facts provided. Do not invent URLs or metrics.
 Return JSON: {"fix": "2-4 sentences with specific steps", "effort": "low|medium|high"}"""
 
+FIX_SUGGESTION_ISSUE_SYSTEM = ISSUE_FIX_SYSTEM
+
+FIX_SUGGESTION_LIGHTHOUSE_SYSTEM = """You are a web performance and Lighthouse audit specialist.
+Given one Lighthouse finding (quick win, diagnostic, or audit), return a concise actionable fix.
+Use ONLY the facts provided. Reference audit IDs and evidence when present. Do not invent URLs or savings.
+Return JSON: {"fix": "2-4 sentences with specific steps (config snippets welcome)", "effort": "low|medium|high"}"""
+
+FIX_SUGGESTION_SECURITY_SYSTEM = """You are a web security and HTTP headers consultant.
+Given one security finding or missing header, return a concise actionable fix with server config guidance when relevant.
+Use ONLY the facts provided. Do not invent URLs or CVEs.
+Return JSON: {"fix": "2-4 sentences with specific steps", "effort": "low|medium|high"}"""
+
+FIX_SUGGESTION_BROWSER_SYSTEM = """You are a frontend debugging specialist.
+Given one browser console error, page exception, or on-page warning, return a concise root-cause fix.
+Use ONLY the facts provided (message, URL, source file, line, stack). Do not invent stack frames.
+Return JSON: {"fix": "2-4 sentences with specific steps", "effort": "low|medium|high"}"""
+
+FIX_SUGGESTION_SEO_CONTENT_SYSTEM = """You are an SEO content strategist.
+Given one content/keyword/structured-data issue (misalignment, cannibalisation, rich results, duplicates, etc.),
+return a concise actionable fix with clear next steps (canonical, merge, redirect, schema, internal links).
+Use ONLY the facts provided. Do not invent traffic numbers.
+Return JSON: {"fix": "2-4 sentences with specific steps", "effort": "low|medium|high"}"""
+
+FIX_SUGGESTION_TECHNICAL_SYSTEM = """You are a technical SEO engineer.
+Given one technical crawl issue (broken link, redirect chain, mixed content, headers, indexing flags),
+return a concise actionable fix.
+Use ONLY the facts provided. Do not invent URLs.
+Return JSON: {"fix": "2-4 sentences with specific steps", "effort": "low|medium|high"}"""
+
+FIX_SUGGESTION_PROMPTS: dict[str, str] = {
+    "issue": FIX_SUGGESTION_ISSUE_SYSTEM,
+    "lighthouse": FIX_SUGGESTION_LIGHTHOUSE_SYSTEM,
+    "security": FIX_SUGGESTION_SECURITY_SYSTEM,
+    "browser": FIX_SUGGESTION_BROWSER_SYSTEM,
+    "seo_content": FIX_SUGGESTION_SEO_CONTENT_SYSTEM,
+    "technical": FIX_SUGGESTION_TECHNICAL_SYSTEM,
+}
+
 AUDIT_EXECUTIVE_SYSTEM = """You write a short executive summary for a site audit report for agency clients.
 Use ONLY the scores and issues provided. Be direct and prioritize by traffic impact.
 Return JSON: {"summary": "3-5 sentences in plain language", "priorities": ["bullet 1", "bullet 2", "bullet 3"]}"""
diff --git a/src/website_profiling/mcp/server.py b/src/website_profiling/mcp/server.py
index bee74233..ca7808a3 100644
--- a/src/website_profiling/mcp/server.py
+++ b/src/website_profiling/mcp/server.py
@@ -89,7 +89,11 @@ def _tools_catalog_json() -> str:
             domains["export"].append(name)
         elif name.startswith(("get_image_", "list_pages_without_lazy", "list_pages_with_images_missing", "list_site_image", "list_lighthouse_image", "list_largest_images", "list_unoptimized_images", "list_images_needing")):
             domains["images"].append(name)
-        elif name.startswith(("list_propert", "get_propert", "get_report", "get_executive", "get_site", "list_report", "get_portfolio")):
+        elif name.startswith(("list_propert", "get_propert", "get_report", "get_executive", "get_site", "list_report", "get_portfolio")) or name in (
+            "get_ads_txt_status",
+            "get_security_txt_status",
+            "get_contact_intelligence",
+        ):
             domains["portfolio"].append(name)
         elif "issue" in name or "category" in name or "workflow" in name:
             domains["issues"].append(name)
@@ -99,7 +103,7 @@ def _tools_catalog_json() -> str:
             domains["schema"].append(name)
         elif "orphan" in name or "link" in name or "fingerprint" in name or "pagerank" in name:
             domains["links"].append(name)
-        elif "indexation" in name or "hreflang" in name or "language" in name:
+        elif "indexation" in name or "hreflang" in name or "language" in name or name == "list_subdomains":
             domains["indexation"].append(name)
         elif "content" in name or "social" in name or "ner" in name or "thin" in name or "opportunit" in name or "duplicate" in name:
             domains["content"].append(name)
diff --git a/src/website_profiling/progress.py b/src/website_profiling/progress.py
new file mode 100644
index 00000000..6776c0f4
--- /dev/null
+++ b/src/website_profiling/progress.py
@@ -0,0 +1,83 @@
+"""Structured pipeline progress events for the web UI (@progress JSON lines)."""
+from __future__ import annotations
+
+import json
+import sys
+import time
+from typing import Any
+
+PREFIX = "@progress "
+
+
+def emit_progress(
+    phase: str,
+    step: str,
+    *,
+    current: int | None = None,
+    total: int | None = None,
+    url: str | None = None,
+    message: str | None = None,
+    elapsed_ms: int | None = None,
+    avg_ms: float | None = None,
+) -> None:
+    """Emit a machine-readable progress line to stdout."""
+    payload: dict[str, Any] = {
+        "phase": phase,
+        "step": step,
+        "ts": int(time.time() * 1000),
+    }
+    if current is not None:
+        payload["current"] = current
+    if total is not None:
+        payload["total"] = total
+    if url:
+        payload["url"] = url
+    if message:
+        payload["message"] = message
+    if elapsed_ms is not None:
+        payload["elapsed_ms"] = elapsed_ms
+    if avg_ms is not None:
+        payload["avg_ms"] = round(avg_ms, 1)
+    print(f"{PREFIX}{json.dumps(payload, ensure_ascii=False)}", flush=True)
+
+
+def emit_phase_start(phase: str, message: str | None = None) -> None:
+    emit_progress(phase, "start", message=message or f"{phase} starting")
+
+
+def emit_phase_done(phase: str, message: str | None = None) -> None:
+    emit_progress(phase, "done", message=message or f"{phase} complete")
+
+
+class CrawlProgressTracker:
+    """Throttle crawl progress emissions (every 2s or every 5 pages)."""
+
+    def __init__(self, total: int | None, start_time: float | None = None) -> None:
+        self.total = total if total and total != float("inf") else None
+        self.start_time = start_time or time.time()
+        self._last_emit = 0.0
+        self._last_count = 0
+        self._last_url: str | None = None
+
+    def maybe_emit(self, current: int, url: str | None = None, *, force: bool = False) -> None:
+        if url:
+            self._last_url = url
+        now = time.time()
+        delta_pages = current - self._last_count
+        is_complete = self.total is not None and current >= self.total
+        if not force and not is_complete:
+            if current > 0 and delta_pages < 5 and (now - self._last_emit) < 2.0:
+                return
+        elapsed_ms = int((now - self.start_time) * 1000)
+        avg_ms = elapsed_ms / current if current > 0 else None
+        emit_progress(
+            "crawl",
+            "fetch",
+            current=current,
+            total=self.total,
+            url=self._last_url,
+            elapsed_ms=elapsed_ms,
+            avg_ms=avg_ms,
+        )
+        self._last_emit = now
+        self._last_count = current
diff --git a/src/website_profiling/reporting/builder.py b/src/website_profiling/reporting/builder.py
index e040bd77..1cff6ed5 100644
--- a/src/website_profiling/reporting/builder.py
+++ b/src/website_profiling/reporting/builder.py
@@ -327,19 +327,30 @@ def fetch(src):
 
 
 def _fetch_site_level(start_url: str, timeout: int = 8) -> dict:
-    """Fetch robots.txt and sitemap.xml from start_url origin. Return site_level dict."""
+    """Fetch robots.txt, sitemap.xml, ads.txt, and security.txt from start_url origin."""
+    from .site_files import fetch_ads_txt, fetch_security_txt, merge_site_file_fields
+
     parsed = urlparse(start_url)
     if not parsed.scheme or not parsed.netloc:
-        return {"robots_present": False, "sitemap_present": False, "sitemap_valid": False}
+        return {
+            "robots_present": False,
+            "sitemap_present": False,
+            "sitemap_valid": False,
+            "ads_txt_present": False,
+            "security_txt_present": False,
+        }
     base = f"{parsed.scheme}://{parsed.netloc}"
     session = requests.Session()
     session.headers.update({"User-Agent": "WebsiteProfiling/1.0"})
-    out = {"robots_present": False, "sitemap_present": False, "sitemap_valid": False}
+    out: dict[str, Any] = {
+        "robots_present": False,
+        "sitemap_present": False,
+        "sitemap_valid": False,
+    }
     try:
         r = session.get(f"{base}/robots.txt", timeout=timeout)
         if r.status_code == 200 and r.text:
             out["robots_present"] = True
-            # Optional: parse robots for sitemap URL
             for line in r.text.splitlines():
                 line = line.strip()
                 if line.lower().startswith("sitemap:"):
@@ -350,10 +361,11 @@ def _fetch_site_level(start_url: str, timeout: int = 8) -> dict:
         r = session.get(f"{base}/sitemap.xml", timeout=timeout)
         if r.status_code == 200 and r.text:
             out["sitemap_present"] = True
-            # Basic XML check
             out["sitemap_valid"] = "<" in r.text and ">" in r.text and ("urlset" in r.text or "sitemapindex" in r.text)
     except Exception:
         pass
+    merge_site_file_fields(out, fetch_ads_txt(session, base, timeout=timeout))
+    merge_site_file_fields(out, fetch_security_txt(session, base, timeout=timeout))
     return out
 
 
@@ -1069,6 +1081,9 @@ def run_simple_report(
     )
     run_id = None
     crawl_run_created_at: Optional[str] = None
+    from ..progress import emit_progress
+
+    emit_progress("report", "load_crawl", message="Loading crawl data from DB")
     print("  Loading crawl data from DB...", flush=True)
     with db_session() as conn:
         run_id = get_latest_crawl_run_id(conn)
@@ -1105,6 +1120,7 @@ def run_simple_report(
     report_display_title = (report_title or "").strip() or f"{site_display} — Crawl Report"
 
     if not edges and not df.empty:
+        emit_progress("report", "build_edges", message="Building edges from crawl data")
         print("  Building edges from crawl data...", flush=True)
         render_mode = (str((config or {}).get("crawl_render_mode") or "static")).strip().lower()
         js_concurrency_cfg = get_int(config or {}, "crawl_js_concurrency", 3) or 3
@@ -1136,12 +1152,19 @@ def run_simple_report(
 
     # Long report work (ML, graph, network) runs without a DB handle; payload write uses db_session again.
 
+    emit_progress("report", "seo_summary", message="Computing SEO summary and issues")
     print("  Computing SEO summary and issues...", flush=True)
     summary_seo = _compute_summary_seo_issues(df)
 
+    emit_progress("report", "site_level", message="Fetching site-level data")
     print("  Fetching site-level (robots.txt, sitemap)...", flush=True)
     site_level = _fetch_site_level(start_url or "", timeout=8)
 
+    emit_progress("report", "contact_intelligence", message="Building contact intelligence")
+    from .contact_intelligence import build_contact_intelligence
+
+    contact_intelligence = build_contact_intelligence(df, site_level, start_url or "", config)
+
     site_ssl_expires_at: Optional[str] = None
     su = (start_url or "").strip()
     if su.lower().startswith("https://"):
@@ -1152,6 +1175,7 @@ def run_simple_report(
 
     security_findings: list = []
     if run_security_scan_flag:
+        emit_progress("report", "security_scan", message="Running security scan")
         print("  Running security scan...", flush=True)
         security_findings = run_security_scan(
             df,
@@ -1163,12 +1187,14 @@ def run_simple_report(
         )
         print(f"  Security scan: {len(security_findings)} findings.", flush=True)
 
+    emit_progress("report", "content_analysis", message="Content analysis")
     print("  Content analysis (crawl + optional AI insights)...", flush=True)
     local_bundle = run_local_enrichment(df, config)
     llm_cfg = load_llm_config_from_db()
     llm_bundle = run_llm_enrichment(df, llm_cfg) if llm_is_enabled(llm_cfg) else {}
     ml_bundle = merge_bundles(local_bundle, llm_bundle)
 
+    emit_progress("report", "categories", message="Building report categories")
     print("  Building report categories...", flush=True)
 
     crux_summary: Optional[dict[str, Any]] = None
@@ -1186,12 +1212,21 @@ def run_simple_report(
         lighthouse_summary=lighthouse_summary,
         ml_bundle=ml_bundle,
         crux_summary=crux_summary,
+        lighthouse_by_url=lighthouse_by_url,
     )
     # Ensure categories are JSON-serializable (score may be None)
     for cat in categories:
         if "score" in cat and cat["score"] is not None and hasattr(cat["score"], "item"):
             cat["score"] = int(cat["score"])
 
+    optional_audit_meta: dict[str, Any] = {}
+    try:
+        from .optional_audits import apply_optional_audits
+
+        optional_audit_meta = apply_optional_audits(categories, df, config)
+    except Exception as e:
+        ml_bundle.setdefault("ml_errors", []).append(f"optional_audits: {e}")
+
     df["status_str"] = df["status"].astype(str) if "status" in df.columns else "unknown"
     status_counts = df["status_str"].value_counts().to_dict()
     df["mime"] = (
@@ -1407,6 +1442,10 @@ def _bool_col(col):
         # Tech stack
         rec["tech_stack"] = _str_col("tech_stack")
 
+        # Custom extraction (regex + XPath/CSS extractors)
+        rec["custom_extract"] = _str_col("custom_extract")
+        rec["custom_fields"] = _str_col("custom_fields")
+
         pa_obj: dict[str, Any] = {}
         if "page_analysis" in df.columns:
             raw_pa = row.get("page_analysis")
@@ -1586,6 +1625,7 @@ def _bool_col(col):
         "missing_dimensions": missing_dimensions,
     }
 
+    emit_progress("report", "content_analytics", message="Building content analytics")
     print("  Building content analytics...", flush=True)
     content_analytics = _build_content_analytics(df)
     semantic_keyword_clusters: list[dict[str, Any]] = []
@@ -1620,6 +1660,7 @@ def _bool_col(col):
         "recommendations": summary_seo["recommendations"],
         "categories": categories,
         "site_level": site_level,
+        "contact_intelligence": contact_intelligence,
         "redirects": summary_seo["issues"].get("redirects", []),
         "orphan_urls": [rec["url"] for rec in links if rec.get("inlinks", 0) == 0],
         "status_counts": status_counts,
@@ -1653,7 +1694,62 @@ def _bool_col(col):
         "url_fingerprints": url_fingerprints,
         "keyword_opportunities": keyword_opportunities,
         "ml_errors": ml_bundle.get("ml_errors") or [],
+        **optional_audit_meta,
     }
+    if get_bool(config or {}, "enable_rich_results_validation", False):
+        try:
+            from ..integrations.google.rich_results import summarize_rich_results, validate_urls
+            from ..config import get_str
+
+            sample_urls = [
+                str(l.get("url") or "")
+                for l in links
+                if isinstance(l, dict) and str(l.get("status") or "").startswith("2")
+            ][:20]
+            links_by_url = {
+                str(l.get("url") or ""): l for l in links if isinstance(l, dict) and l.get("url")
+            }
+            creds = None
+            property_id_rr: Optional[int] = None
+            gsc_site = (get_str(config or {}, "gsc_site_url", "") or "").strip() or None
+            try:
+                from ..db import db_session as _rr_db
+                from ..commands.config_resolve import resolve_property_id_from_cfg
+
+                with _rr_db() as conn:
+                    property_id_rr = resolve_property_id_from_cfg(config, conn)
+            except Exception:
+                property_id_rr = None
+            if property_id_rr:
+                try:
+                    from ..integrations.google.auth import build_credentials
+
+                    creds = build_credentials(property_id_rr)
+                except Exception:
+                    creds = None
+            rr_api_key = (get_str(config or {}, "google_rich_results_api_key", "") or "").strip() or None
+            rr_rows = validate_urls(
+                sample_urls,
+                api_key=rr_api_key,
+                creds=creds,
+                site_url=gsc_site,
+                links_by_url=links_by_url,
+            )
+            report_data["rich_results_validation"] = rr_rows
+            report_data["rich_results_meta"] = summarize_rich_results(rr_rows)
+        except Exception as e:
+            report_data.setdefault("ml_errors", []).append(f"rich_results: {e}")
+    try:
+        from ..config import get_str
+        import json as _json
+
+        comp_kw = (get_str(config or {}, "competitor_keyword_gap_json", "") or "").strip()
+        if comp_kw:
+            parsed = _json.loads(comp_kw)
+            if isinstance(parsed, list):
+                report_data["competitor_keyword_gap"] = parsed
+    except Exception as e:
+        report_data.setdefault("ml_errors", []).append(f"competitor_keywords: {e}")
     if run_id is not None:
         report_data["crawl_run_id"] = run_id
         report_data["crawl_run_created_at"] = crawl_run_created_at
@@ -1662,6 +1758,7 @@ def _bool_col(col):
         report_data["lighthouse_diagnostics"] = lighthouse_summary.get("diagnostics") or []
         report_data["lighthouse_human_summary"] = lighthouse_summary.get("human_summary_full") or lighthouse_summary.get("human_summary") or ""
     report_data["lighthouse_by_url"] = lighthouse_by_url
+    emit_progress("report", "write_payload", message="Writing report payload to DB")
     print("  Writing report payload to DB...", flush=True)
     from ..db import db_session as _db, write_report_payload as db_write_report_payload
     with _db() as conn:
@@ -1681,8 +1778,27 @@ def _bool_col(col):
             google_data = read_latest_google_data(conn, property_id=property_id)
             if google_data:
                 report_data["google"] = google_data
+                from .issue_impact import enrich_categories_with_traffic_impact
+
+                enrich_categories_with_traffic_impact(
+                    report_data.get("categories") or [],
+                    google_data,
+                )
         except Exception:
             pass
+        try:
+            from ..db.crawl_store import read_link_edges
+            from .link_edges_report import build_inlink_anchor_matrix, summarize_link_rel
+
+            emit_progress("report", "link_edges", message="Loading crawl link edges")
+            if run_id is not None:
+                link_edges = read_link_edges(conn, run_id, limit=15000)
+                if link_edges:
+                    report_data["link_edges"] = link_edges
+                    report_data["link_rel_summary"] = summarize_link_rel(link_edges)
+                    report_data["inlink_anchor_matrix"] = build_inlink_anchor_matrix(link_edges)
+        except Exception as e:
+            report_data.setdefault("ml_errors", []).append(f"link_edges: {e}")
         try:
             from ..integrations.google.keyword_store import read_latest_keyword_data
             from ..integrations.google.gsc_links_store import read_latest_gsc_links_data
@@ -1720,6 +1836,15 @@ def _bool_col(col):
             from .categories import merge_indexation_issues
 
             merge_indexation_issues(report_data.get("categories") or [], df, indexation_cov)
+            emit_progress("report", "subdomains", message="Building subdomain inventory")
+            from .subdomains import build_subdomain_inventory
+            from .categories import merge_subdomain_issues
+
+            subdomains_data = build_subdomain_inventory(df, indexation_cov, start_url or "", config)
+            report_data["subdomains"] = subdomains_data
+            if subdomains_data.get("crtsh_error"):
+                report_data.setdefault("ml_errors", []).append(subdomains_data["crtsh_error"])
+            merge_subdomain_issues(report_data.get("categories") or [], subdomains_data)
         except Exception as e:
             report_data.setdefault("ml_errors", []).append(f"indexation: {e}")
         try:
@@ -1769,6 +1894,48 @@ def _bool_col(col):
             report_data["executive_summary"] = generate_audit_executive_summary(report_data, config)
         except Exception:
             pass
+        try:
+            from ..tools.audit_tools.llm_tools import get_portfolio_summary
+            from ..tools.audit_tools.context import AuditToolContext
+
+            portfolio = get_portfolio_summary(conn, AuditToolContext(property_id=property_id), {})
+            scores = []
+            for c in report_data.get("categories") or []:
+                try:
+                    if c.get("score") is not None:
+                        scores.append(int(float(c.get("score"))))
+                except (TypeError, ValueError):
+                    continue
+            prop_health = round(sum(scores) / len(scores)) if scores else None
+            prop_count = int(portfolio.get("count") or 0)
+            median = portfolio.get("median_health_score")
+            bench: dict[str, Any] = {
+                "median_health_score": median,
+                "property_health_score": prop_health,
+                "property_count": prop_count,
+            }
+            if prop_count <= 1:
+                bench["status"] = "single_property"
+                bench["message"] = "Add more properties to compare portfolio median."
+            elif median is None:
+                bench["status"] = "unavailable"
+                bench["message"] = "No health snapshots yet for portfolio comparison."
+            else:
+                bench["status"] = "ok"
+            report_data["portfolio_benchmark"] = bench
+            print(
+                f"  Portfolio benchmark: property={prop_health}, median={median}, count={prop_count}",
+                flush=True,
+            )
+        except Exception as e:
+            report_data.setdefault("ml_errors", []).append(f"portfolio_benchmark: {e}")
+            report_data["portfolio_benchmark"] = {
+                "status": "error",
+                "message": str(e),
+                "property_health_score": None,
+                "median_health_score": None,
+                "property_count": 0,
+            }
         report_data["report_meta"] = _build_report_metadata(
             df,
             config,
diff --git a/src/website_profiling/reporting/categories.py b/src/website_profiling/reporting/categories.py
index 67c81ec3..15014e7d 100644
--- a/src/website_profiling/reporting/categories.py
+++ b/src/website_profiling/reporting/categories.py
@@ -2,6 +2,7 @@
 Report categories for site audits: Technical SEO, Core Web Vitals, Performance,
 Accessibility & markup, Links, Mobile SEO, Security, Content quality.
 """
+import json
 from typing import Any, Optional
 from urllib.parse import urlparse
 
@@ -207,6 +208,29 @@ def merge_indexation_issues(categories: list[dict], df: pd.DataFrame, indexation
             break
 
 
+def merge_subdomain_issues(categories: list[dict], subdomains: dict | None) -> None:
+    """Append GSC subdomain gap summary to technical SEO."""
+    if not subdomains or subdomains.get("disabled"):
+        return
+    hosts = subdomains.get("gsc_hosts_not_crawled") or []
+    if not hosts:
+        return
+    preview = ", ".join(hosts[:5])
+    suffix = f" (+{len(hosts) - 5} more)" if len(hosts) > 5 else ""
+    msg = f"GSC shows URLs on subdomain(s) not reached by crawl: {preview}{suffix}."
+    issue = _issue(
+        msg,
+        priority="Medium",
+        recommendation="Include these hosts in crawl scope or verify they are intentional separate properties.",
+    )
+    for cat in categories:
+        if cat.get("id") == "technical_seo":
+            cat["issues"] = _sort_issues((cat.get("issues") or []) + [issue])
+            recs = {i["recommendation"] for i in cat["issues"] if i.get("recommendation")}
+            cat["recommendations"] = list(recs)
+            break
+
+
 def _orphan_hub_suggestions(edges: list[tuple[str, str]], orphan_urls: list[str]) -> list[dict]:
     issues: list[dict] = []
     if not edges or not orphan_urls:
@@ -259,6 +283,18 @@ def category_technical_seo(
             recommendation="Ensure sitemap is valid XML and follows sitemaps.org format.",
         ))
         deductions.append((5, True))
+    if site_level.get("ads_txt_present") is False:
+        issues.append(_issue(
+            "ads.txt is missing or unreachable.",
+            priority="Low",
+            recommendation="Add an ads.txt file at the site root if you run programmatic advertising.",
+        ))
+    if site_level.get("security_txt_present") is False:
+        issues.append(_issue(
+            "security.txt is missing or unreachable.",
+            priority="Low",
+            recommendation="Publish security.txt at /.well-known/security.txt with a Contact field for security reporting.",
+        ))
 
     # Canonical: missing or self-mismatch
     if "canonical_url" in df.columns and len(success_df) > 0:
@@ -544,8 +580,84 @@ def category_performance(df: pd.DataFrame) -> dict:
     }
 
 
-def category_html_accessibility(df: pd.DataFrame) -> dict:
-    """HTML and Accessibility: semantic HTML, heading structure, alt, ARIA, contrast (stub)."""
+def _parse_page_analysis_cell(raw: object) -> dict[str, Any]:
+    if isinstance(raw, dict):
+        return raw
+    if not raw or not isinstance(raw, str):
+        return {}
+    try:
+        parsed = json.loads(raw)
+        return parsed if isinstance(parsed, dict) else {}
+    except Exception:
+        return {}
+
+
+def contrast_issues_from_sources(
+    df: pd.DataFrame,
+    lighthouse_by_url: Optional[dict[str, Any]] = None,
+) -> list[dict]:
+    """Contrast issues from axe crawl data and per-URL Lighthouse failures."""
+    issues: list[dict] = []
+    seen_urls: set[str] = set()
+
+    if df is not None and not df.empty and "page_analysis" in df.columns:
+        for _, row in df.iterrows():
+            url = str(row.get("url") or "").strip()
+            if not url:
+                continue
+            pa = _parse_page_analysis_cell(row.get("page_analysis"))
+            axe = pa.get("axe_violations")
+            if not isinstance(axe, list):
+                continue
+            contrast_hits = [
+                v for v in axe
+                if isinstance(v, dict) and "color-contrast" in str(v.get("id") or "")
+            ]
+            if not contrast_hits:
+                continue
+            seen_urls.add(url.rstrip("/"))
+            first = contrast_hits[0]
+            msg = str(first.get("description") or first.get("help") or "Color contrast violation")
+            issues.append(_issue(
+                f"axe: {msg}",
+                url=url,
+                priority="Medium",
+                recommendation=str(
+                    first.get("help")
+                    or "Fix text/background contrast to meet WCAG AA (axe-core)."
+                ),
+            ))
+
+    lh_map = lighthouse_by_url or {}
+    for url, summary in lh_map.items():
+        if not isinstance(summary, dict):
+            continue
+        u = str(url or summary.get("url") or "").strip().rstrip("/")
+        if not u or u in seen_urls:
+            continue
+        for fail in summary.get("top_failures") or []:
+            if not isinstance(fail, dict):
+                continue
+            if str(fail.get("id") or "") != "color-contrast":
+                continue
+            seen_urls.add(u)
+            help_text = str(fail.get("helpText") or "Low color contrast")
+            issues.append(_issue(
+                f"Lighthouse: {help_text}",
+                url=u,
+                priority="Medium",
+                recommendation="Increase contrast ratio between text and background to meet WCAG AA.",
+            ))
+            break
+
+    return issues[:40]
+
+
+def category_html_accessibility(
+    df: pd.DataFrame,
+    lighthouse_by_url: Optional[dict[str, Any]] = None,
+) -> dict:
+    """HTML and Accessibility: semantic HTML, heading structure, alt, ARIA, contrast."""
     issues = []
     deductions = []
     success_df = df[df["status"].astype(str).str.match(r"2\d{2}", na=False)] if "status" in df.columns else pd.DataFrame()
@@ -628,11 +740,16 @@ def category_html_accessibility(df: pd.DataFrame) -> dict:
             ))
             deductions.append((min(10, complex_pages * 2), True))
 
-    issues.append(_issue(
-        "Color contrast is not measured by this tool.",
-        priority="Low",
-        recommendation="Use browser DevTools or axe to check contrast and accessibility.",
-    ))
+    contrast_issues = contrast_issues_from_sources(df, lighthouse_by_url)
+    if contrast_issues:
+        issues.extend(contrast_issues)
+        deductions.append((min(25, len(contrast_issues) * 4), True))
+    else:
+        issues.append(_issue(
+            "Color contrast is not measured by this tool.",
+            priority="Low",
+            recommendation="Enable axe (browser crawl) or Lighthouse to check contrast.",
+        ))
 
     score = _score_deductions(100, deductions)
     if len(success_df) > 0 and score == 0:
@@ -905,6 +1022,7 @@ def build_categories(
     lighthouse_summary: Optional[dict] = None,
     ml_bundle: Optional[dict] = None,
     crux_summary: Optional[dict] = None,
+    lighthouse_by_url: Optional[dict[str, Any]] = None,
 ) -> list[dict]:
     """
     Build all category dicts with score, issues (with priority and recommendation), and recommendations.
@@ -926,7 +1044,7 @@ def build_categories(
         category_technical_seo(df, site_level),
         cwv,
         category_performance(df),
-        category_html_accessibility(df),
+        category_html_accessibility(df, lighthouse_by_url=lighthouse_by_url),
         category_link_health(df, edges, issues_broken, issues_redirects),
         category_mobile(df),
         category_security(df, site_level, start_url or "", security_findings=security_findings),
diff --git a/src/website_profiling/reporting/contact_intelligence.py b/src/website_profiling/reporting/contact_intelligence.py
new file mode 100644
index 00000000..19d3042f
--- /dev/null
+++ b/src/website_profiling/reporting/contact_intelligence.py
@@ -0,0 +1,173 @@
+"""Aggregate business contact signals from crawl, site files, and RDAP."""
+from __future__ import annotations
+
+import json
+import re
+from typing import Any, Optional
+from urllib.parse import urlparse
+
+import pandas as pd
+
+from ..config import get_bool
+from .site_files import fetch_rdap_org_name
+
+_CONTACT_PAGE_RE = re.compile(r"/(contact|about|support)(/|$)", re.I)
+_ORG_SCHEMA_TYPES = frozenset({"organization", "localbusiness", "corporation"})
+_LIST_LIMIT = 50
+
+
+def _parse_page_analysis(raw: object) -> dict[str, Any]:
+    if isinstance(raw, dict):
+        return raw
+    if not raw or (isinstance(raw, float) and pd.isna(raw)):
+        return {}
+    s = str(raw).strip()
+    if not s or s == "{}":
+        return {}
+    try:
+        parsed = json.loads(s)
+        return parsed if isinstance(parsed, dict) else {}
+    except json.JSONDecodeError:
+        return {}
+
+
+def _merge_entry(
+    bucket: dict[str, dict[str, Any]],
+    value: str,
+    *,
+    source: str,
+    url: str = "",
+) -> None:
+    val = value.strip()
+    if not val:
+        return
+    key = val.lower()
+    if key not in bucket:
+        bucket[key] = {"value": val, "sources": [], "urls": []}
+    entry = bucket[key]
+    if source not in entry["sources"]:
+        entry["sources"].append(source)
+    if url and url not in entry["urls"]:
+        entry["urls"].append(url)
+
+
+def _signals_from_page(pa: dict[str, Any]) -> dict[str, list[str]]:
+    signals = pa.get("contact_signals")
+    if not isinstance(signals, dict):
+        return {}
+    out: dict[str, list[str]] = {}
+    for key in ("emails", "phones", "addresses", "organization_names"):
+        raw = signals.get(key)
+        if isinstance(raw, list):
+            out[key] = [str(x).strip() for x in raw if str(x).strip()]
+    return out
+
+
+def _has_org_schema(pa: dict[str, Any]) -> bool:
+    types = pa.get("json_ld_types") or []
+    if isinstance(types, str):
+        types = [types]
+    if not isinstance(types, list):
+        return False
+    return any(str(t).strip().lower() in _ORG_SCHEMA_TYPES for t in types)
+
+
+def _apex_from_start_url(start_url: str) -> str:
+    parsed = urlparse((start_url or "").strip())
+    host = (parsed.netloc or "").lower()
+    if host.startswith("www."):
+        host = host[4:]
+    return host
+
+
+def build_contact_intelligence(
+    df: pd.DataFrame,
+    site_level: dict[str, Any],
+    start_url: str,
+    config: dict[str, str] | None = None,
+) -> dict[str, Any]:
+    """Build sourced contact intelligence from crawl page_analysis and site files."""
+    emails: dict[str, dict[str, Any]] = {}
+    phones: dict[str, dict[str, Any]] = {}
+    addresses: dict[str, dict[str, Any]] = {}
+    org_names: dict[str, dict[str, Any]] = {}
+
+    success_df = (
+        df[df["status"].astype(str).str.match(r"2\d{2}", na=False)]
+        if "status" in df.columns and not df.empty
+        else pd.DataFrame()
+    )
+    page_scores: list[tuple[int, str]] = []
+    has_org_on_home = False
+
+    for _, row in success_df.iterrows():
+        url = str(row.get("url") or "").strip()
+        if not url:
+            continue
+        pa = _parse_page_analysis(row.get("page_analysis"))
+        signals = _signals_from_page(pa)
+        org_schema = _has_org_schema(pa)
+        path = urlparse(url).path or "/"
+        if path in ("/", "") and org_schema:
+            has_org_on_home = True
+
+        score = sum(len(signals.get(k) or []) for k in signals)
+        if score:
+            page_scores.append((score, url))
+
+        for email in signals.get("emails") or []:
+            _merge_entry(emails, email, source="json_ld" if org_schema else "crawl", url=url)
+        for phone in signals.get("phones") or []:
+            _merge_entry(phones, phone, source="tel_link", url=url)
+        for addr in signals.get("addresses") or []:
+            _merge_entry(addresses, addr, source="json_ld", url=url)
+        for org in signals.get("organization_names") or []:
+            _merge_entry(org_names, org, source="json_ld", url=url)
+
+    for contact in site_level.get("security_txt_contact") or []:
+        if not isinstance(contact, str):
+            continue
+        c = contact.strip()
+        if c.lower().startswith("mailto:"):
+            _merge_entry(emails, c[7:].split("?")[0], source="security_txt", url="")
+        elif c.lower().startswith("tel:"):
+            _merge_entry(phones, c[4:].split("?")[0], source="security_txt", url="")
+        elif "@" in c:
+            _merge_entry(emails, c, source="security_txt", url="")
+
+    if get_bool(config or {}, "enable_rdap_org_lookup", True):
+        apex = _apex_from_start_url(start_url)
+        if apex:
+            org = fetch_rdap_org_name(apex)
+            if org:
+                _merge_entry(org_names, org, source="whois_org", url="")
+
+    primary_contact_page: Optional[str] = None
+    if page_scores:
+        page_scores.sort(key=lambda x: (-x[0], x[1]))
+        for _, candidate in page_scores:
+            if _CONTACT_PAGE_RE.search(urlparse(candidate).path or ""):
+                primary_contact_page = candidate
+                break
+        if not primary_contact_page:
+            primary_contact_page = page_scores[0][1]
+
+    consistency_notes: list[str] = []
+    if len(emails) > 3:
+        consistency_notes.append(f"{len(emails)} distinct email addresses found across the site.")
+    if len(org_names) > 1:
+        consistency_notes.append(f"{len(org_names)} distinct organization names found in structured data.")
+    if not has_org_on_home:
+        consistency_notes.append("No Organization (or LocalBusiness) schema detected on the homepage.")
+
+    def _list_items(bucket: dict[str, dict[str, Any]], limit: int = _LIST_LIMIT) -> list[dict[str, Any]]:
+        return list(bucket.values())[:limit]
+
+    return {
+        "emails": _list_items(emails),
+        "phones": _list_items(phones),
+        "addresses": _list_items(addresses),
+        "organization_names": _list_items(org_names),
+        "primary_contact_page": primary_contact_page,
+        "consistency_notes": consistency_notes,
+    }
diff --git a/src/website_profiling/reporting/issue_impact.py b/src/website_profiling/reporting/issue_impact.py
new file mode 100644
index 00000000..c142af21
--- /dev/null
+++ b/src/website_profiling/reporting/issue_impact.py
@@ -0,0 +1,85 @@
+"""Attach GSC/GA4 traffic signals and impact scores to audit issues."""
+from __future__ import annotations
+
+from typing import Any
+
+PRIORITY_WEIGHT = {"Critical": 1000, "High": 100, "Medium": 10, "Low": 1}
+
+
+def _metrics_by_url(google_data: dict[str, Any] | None) -> tuple[dict[str, dict], dict[str, dict]]:
+    clicks: dict[str, dict] = {}
+    sessions: dict[str, dict] = {}
+    if not google_data or not isinstance(google_data, dict):
+        return clicks, sessions
+    gsc = google_data.get("gsc") or {}
+    if isinstance(gsc, dict):
+        for row in (gsc.get("pages") or gsc.get("top_pages") or []):
+            if not isinstance(row, dict):
+                continue
+            url = str(row.get("page") or row.get("url") or "").strip().lower().rstrip("/")
+            if not url:
+                continue
+            clicks[url] = {
+                "gsc_clicks": float(row.get("clicks") or 0),
+                "gsc_impressions": float(row.get("impressions") or 0),
+            }
+    ga4 = google_data.get("ga4") or {}
+    if isinstance(ga4, dict):
+        for row in (ga4.get("pages") or ga4.get("top_pages") or []):
+            if not isinstance(row, dict):
+                continue
+            path = str(row.get("path") or row.get("pagePath") or row.get("url") or "").strip().lower()
+            if not path:
+                continue
+            sessions[path] = {"ga4_sessions": float(row.get("sessions") or 0)}
+    return clicks, sessions
+
+
+def compute_impact_score(
+    priority: str,
+    *,
+    gsc_clicks: float = 0,
+    ga4_sessions: float = 0,
+) -> float:
+    base = PRIORITY_WEIGHT.get(priority, 1)
+    return round(base + gsc_clicks * 10.0 + ga4_sessions * 5.0, 2)
+
+
+def enrich_categories_with_traffic_impact(
+    categories: list[dict[str, Any]],
+    google_data: dict[str, Any] | None,
+) -> list[dict[str, Any]]:
+    """Mutates issue dicts in categories with traffic fields and impact_score."""
+    clicks_by_url, sessions_by_path = _metrics_by_url(google_data)
+    for cat in categories or []:
+        if not isinstance(cat, dict):
+            continue
+        for issue in cat.get("issues") or []:
+            if not isinstance(issue, dict):
+                continue
+            url = str(issue.get("url") or "").strip().lower().rstrip("/")
+            gsc = clicks_by_url.get(url, {})
+            ga4_sess = 0.0
+            if url:
+                for path_key, ga in sessions_by_path.items():
+                    if url.endswith(path_key.rstrip("/")) or path_key in url:
+                        ga4_sess = max(ga4_sess, float(ga.get("ga4_sessions") or 0))
+            issue["gsc_clicks"] = gsc.get("gsc_clicks", 0)
+            issue["gsc_impressions"] = gsc.get("gsc_impressions", 0)
+            issue["ga4_sessions"] = ga4_sess
+            issue["impact_score"] = compute_impact_score(
+                str(issue.get("priority") or "Medium"),
+                gsc_clicks=float(issue.get("gsc_clicks") or 0),
+                ga4_sessions=ga4_sess,
+            )
+    return categories
+
+
+def sort_issues_by_impact(issues: list[dict[str, Any]]) -> list[dict[str, Any]]:
+    return sorted(
+        issues,
+        key=lambda i: (
+            -float(i.get("impact_score") or 0),
+            PRIORITY_WEIGHT.get(str(i.get("priority") or "Low"), 99),
+        ),
+    )
diff --git a/src/website_profiling/reporting/link_edges_report.py b/src/website_profiling/reporting/link_edges_report.py
new file mode 100644
index 00000000..6bc98a91
--- /dev/null
+++ b/src/website_profiling/reporting/link_edges_report.py
@@ -0,0 +1,37 @@
+"""Summarize rich link_edges for reports and tools."""
+from __future__ import annotations
+
+from collections import Counter
+from typing import Any
+
+
+def summarize_link_rel(edges: list[dict[str, Any]]) -> dict[str, Any]:
+    internal = [e for e in edges if str(e.get("link_type") or "") == "internal"]
+    return {
+        "total_edges": len(edges),
+        "internal_edges": len(internal),
+        "nofollow_internal": sum(1 for e in internal if e.get("is_nofollow")),
+        "sponsored_internal": sum(1 for e in internal if e.get("is_sponsored")),
+        "ugc_internal": sum(1 for e in internal if e.get("is_ugc")),
+        "external_edges": len(edges) - len(internal),
+    }
+
+
+def build_inlink_anchor_matrix(edges: list[dict[str, Any]], *, limit: int = 500) -> list[dict[str, Any]]:
+    """Aggregate inlink anchor text counts per target URL."""
+    buckets: dict[tuple[str, str], int] = Counter()
+    for e in edges:
+        if str(e.get("link_type") or "") != "internal":
+            continue
+        target = str(e.get("to_url") or "").rstrip("/")
+        anchor = str(e.get("anchor_text") or "").strip() or "(empty)"
+        source = str(e.get("from_url") or "").rstrip("/")
+        if not target or not source:
+            continue
+        buckets[(target, anchor)] += 1
+    rows = [
+        {"target_url": t, "anchor_text": a, "inlink_count": c}
+        for (t, a), c in buckets.items()
+    ]
+    rows.sort(key=lambda r: (-r["inlink_count"], r["target_url"]))
+    return rows[: max(1, limit)]
diff --git a/src/website_profiling/reporting/optional_audits.py b/src/website_profiling/reporting/optional_audits.py
new file mode 100644
index 00000000..3867cb72
--- /dev/null
+++ b/src/website_profiling/reporting/optional_audits.py
@@ -0,0 +1,355 @@
+"""Optional crawl audits: pagination, spell, HTML, AMP, Wayback, axe issues."""
+from __future__ import annotations
+
+import json
+import re
+import sys
+from typing import Any, Optional
+from urllib.parse import urlparse
+
+import pandas as pd
+import requests
+
+from ..config import get_bool
+from ..progress import emit_progress
+from .categories import _issue, _sort_issues
+
+_WAYBACK_CACHE: dict[str, bool] = {}
+
+
+def _parse_page_analysis(raw: object) -> dict[str, Any]:
+    if isinstance(raw, dict):
+        return raw
+    if not raw or not isinstance(raw, str):
+        return {}
+    try:
+        parsed = json.loads(raw)
+        return parsed if isinstance(parsed, dict) else {}
+    except Exception:
+        return {}
+
+
+def _content_quality_category(categories: list[dict]) -> dict | None:
+    for cat in categories:
+        if str(cat.get("id") or "") == "intelligence":
+            return cat
+    return None
+
+
+def _technical_category(categories: list[dict]) -> dict | None:
+    for cat in categories:
+        if str(cat.get("id") or "") == "technical_seo":
+            return cat
+    return None
+
+
+def _accessibility_category(categories: list[dict]) -> dict | None:
+    for cat in categories:
+        if str(cat.get("id") or "") == "html_accessibility":
+            return cat
+    return None
+
+
+def pagination_issues(df: pd.DataFrame) -> list[dict]:
+    issues: list[dict] = []
+    if df is None or df.empty or "page_analysis" not in df.columns:
+        return issues
+    orphan_prev = 0
+    amp_mismatch = 0
+    for _, row in df.iterrows():
+        url = str(row.get("url") or "").strip()
+        if not url:
+            continue
+        pa = _parse_page_analysis(row.get("page_analysis"))
+        pag = pa.get("pagination") if isinstance(pa.get("pagination"), dict) else {}
+        rel_next = pag.get("rel_next")
+        rel_prev = pag.get("rel_prev")
+        if rel_prev and not rel_next:
+            orphan_prev += 1
+        amphtml = pag.get("amphtml")
+        if amphtml and str(row.get("canonical_url") or "").strip() and amphtml != row.get("canonical_url"):
+            amp_mismatch += 1
+    if orphan_prev:
+        issues.append(_issue(
+            f"{orphan_prev} page(s) have rel=prev without rel=next (pagination chain may be broken).",
+            priority="Medium",
+            recommendation="Ensure paginated series use paired rel=next and rel=prev links.",
+        ))
+    if amp_mismatch:
+        issues.append(_issue(
+            f"{amp_mismatch} page(s) have amphtml link that does not match canonical.",
+            priority="Medium",
+            recommendation="Pair AMP and canonical URLs correctly for mobile/desktop variants.",
+        ))
+    return issues
+
+
+def _spell_text_parts(row: pd.Series) -> str:
+    parts = [
+        str(row.get("title") or ""),
+        str(row.get("h1") or ""),
+        str(row.get("content_excerpt") or ""),
+        str(row.get("meta_description") or ""),
+    ]
+    return " ".join(p for p in parts if p.strip())
+
+
+def spell_check_issues(df: pd.DataFrame, *, max_pages: int = 50) -> tuple[list[dict], Optional[str]]:
+    """Return (issues, skip_reason). skip_reason set when pyspellchecker is missing."""
+    issues: list[dict] = []
+    try:
+        from spellchecker import SpellChecker  # type: ignore[import-untyped]
+    except ImportError:
+        return issues, "pyspellchecker not installed (pip install -r requirements-optional.txt)"
+    spell = SpellChecker()
+    checked = 0
+    for _, row in df.iterrows():
+        if checked >= max_pages:
+            break
+        status = str(row.get("status") or "")
+        if not status.startswith("2"):
+            continue
+        excerpt = _spell_text_parts(row).strip()
+        if len(excerpt) < 40:
+            continue
+        words = re.findall(r"[a-zA-Z']{4,}", excerpt.lower())
+        if not words:
+            continue
+        unknown = spell.unknown(words[:120])
+        if len(unknown) >= 3:
+            url = str(row.get("url") or "")
+            sample = ", ".join(sorted(unknown)[:5])
+            issues.append(_issue(
+                f"Possible spelling issues ({sample}).",
+                url=url,
+                priority="Low",
+                recommendation="Review title, H1, and visible copy for typos.",
+            ))
+            checked += 1
+    return issues[:20], None
+
+
+def html_validation_issues(df: pd.DataFrame, *, max_pages: int = 30) -> tuple[list[dict], bool]:
+    """Return issues and whether html5lib parser was used."""
+    issues: list[dict] = []
+    use_parser = False
+    try:
+        import html5lib  # type: ignore[import-untyped]  # noqa: F401
+        use_parser = True
+    except ImportError:
+        use_parser = False
+
+    checked = 0
+    for _, row in df.iterrows():
+        if checked >= max_pages:
+            break
+        html = str(row.get("html") or "")
+        if len(html) < 100:
+            continue
+        url = str(row.get("url") or "")
+        warnings: list[str] = []
+        if use_parser:
+            try:
+                from html5lib import HTMLParser  # type: ignore[import-untyped]
+
+                parser = HTMLParser()
+                parser.parse(html)
+            except Exception as exc:
+                warnings.append(f"parser error: {exc}")
+        if html.count("<title") > 1:
+            warnings.append("multiple title tags")
+        if re.search(r"<html[^>]*>", html, re.I) and not re.search(r"</html>", html, re.I):
+            warnings.append("missing closing html tag")
+        ids = re.findall(r'\bid=["\']([^"\']+)["\']', html, re.I)
+        if len(ids) != len(set(ids)):
+            warnings.append("duplicate id attributes")
+        if warnings:
+            issues.append(_issue(
+                f"HTML structure warnings: {', '.join(warnings)}.",
+                url=url,
+                priority="Low",
+                recommendation="Fix markup validation issues that may affect parsing or accessibility.",
+            ))
+            checked += 1
+    return issues, use_parser
+
+
+def amp_audit_issues(df: pd.DataFrame) -> list[dict]:
+    issues: list[dict] = []
+    if df is None or df.empty:
+        return issues
+    for _, row in df.iterrows():
+        url = str(row.get("url") or "")
+        pa = _parse_page_analysis(row.get("page_analysis"))
+        pag = pa.get("pagination") if isinstance(pa.get("pagination"), dict) else {}
+        amphtml = pag.get("amphtml")
+        is_amp = "/amp" in urlparse(url).path.lower() or bool(
+            re.search(r"\bamp\b", str(row.get("content_type") or ""), re.I)
+        )
+        if not amphtml and not is_amp:
+            continue
+        canon = str(row.get("canonical_url") or "").strip()
+        if not canon:
+            issues.append(_issue(
+                "AMP or amphtml variant missing canonical URL.",
+                url=url,
+                priority="Medium",
+                recommendation="Add canonical link pointing to the preferred non-AMP URL.",
+            ))
+        elif amphtml and canon and amphtml.rstrip("/") != canon.rstrip("/") and is_amp:
+            issues.append(_issue(
+                "AMP page canonical does not match linked amphtml href.",
+                url=url,
+                priority="Medium",
+                recommendation="Align canonical URL with amphtml pairing for AMP variants.",
+            ))
+    return issues[:25]
+
+
+def wayback_issues(df: pd.DataFrame, *, max_lookups: int = 15) -> list[dict]:
+    issues: list[dict] = []
+    if df is None or df.empty:
+        return issues
+    looked = 0
+    for _, row in df.iterrows():
+        if looked >= max_lookups:
+            break
+        status = str(row.get("status") or "")
+        if not status.startswith("404"):
+            continue
+        url = str(row.get("url") or "").strip()
+        if not url:
+            continue
+        cache_key = url.rstrip("/")
+        if cache_key in _WAYBACK_CACHE:
+            if _WAYBACK_CACHE[cache_key]:
+                issues.append(_issue(
+                    "404 URL has Wayback snapshot (Estimated).",
+                    url=url,
+                    priority="Low",
+                    recommendation="Review whether redirect or content restoration is appropriate.",
+                ))
+                looked += 1
+            continue
+        try:
+            resp = requests.get(
+                "https://archive.org/wayback/available",
+                params={"url": url},
+                timeout=8,
+            )
+            data = resp.json()
+            snap = (data.get("archived_snapshots") or {}).get("closest") or {}
+            available = bool(snap.get("available"))
+            _WAYBACK_CACHE[cache_key] = available
+            if available:
+                ts = snap.get("timestamp") or "unknown"
+                issues.append(_issue(
+                    f"404 URL has Wayback snapshot (Estimated, captured {ts}).",
+                    url=url,
+                    priority="Low",
+                    recommendation="Review whether redirect or content restoration is appropriate.",
+                ))
+                looked += 1
+        except Exception:
+            _WAYBACK_CACHE[cache_key] = False
+            continue
+    return issues
+
+
+def axe_issues_from_df(df: pd.DataFrame) -> list[dict]:
+    issues: list[dict] = []
+    if df is None or df.empty or "page_analysis" not in df.columns:
+        return issues
+    for _, row in df.iterrows():
+        pa = _parse_page_analysis(row.get("page_analysis"))
+        axe = pa.get("axe_violations")
+        if not isinstance(axe, list) or not axe:
+            continue
+        url = str(row.get("url") or "")
+        for v in axe[:5]:
+            if not isinstance(v, dict):
+                continue
+            msg = str(v.get("description") or v.get("id") or "axe violation")
+            issues.append(_issue(
+                f"axe: {msg}",
+                url=url,
+                priority="Medium",
+                recommendation=str(v.get("help") or "Fix accessibility violation reported by axe-core."),
+            ))
+    return issues[:40]
+
+
+def apply_optional_audits(
+    categories: list[dict],
+    df: pd.DataFrame,
+    config: Optional[dict[str, str]],
+) -> dict[str, Any]:
+    """Merge optional audit issues into categories; return sidecar payload keys."""
+    cfg = config or {}
+    extras: dict[str, Any] = {}
+    tech = _technical_category(categories)
+    content = _content_quality_category(categories)
+    a11y = _accessibility_category(categories)
+
+    emit_progress("optional", "pagination", message="Checking pagination links")
+    pag = pagination_issues(df)
+    if pag and tech is not None:
+        tech.setdefault("issues", []).extend(pag)
+        tech["issues"] = _sort_issues(tech["issues"])
+        extras["pagination_issues"] = len(pag)
+
+    if get_bool(cfg, "enable_spell_check", False) and content is not None:
+        emit_progress("optional", "spell_check", message="Spell-checking excerpts")
+        spell, skip = spell_check_issues(df)
+        if skip:
+            extras["spell_check_skipped"] = skip
+            print(f"Warning: {skip}", file=sys.stderr, flush=True)
+        if spell:
+            content.setdefault("issues", []).extend(spell)
+            content["issues"] = _sort_issues(content["issues"])
+            extras["spell_check_pages"] = len(spell)
+
+    if get_bool(cfg, "enable_html_validation", False) and tech is not None:
+        emit_progress("optional", "html_validation", message="Validating HTML structure")
+        html_issues, used_parser = html_validation_issues(df)
+        extras["html_validation_parser"] = "html5lib" if used_parser else "regex"
+        if html_issues:
+            tech.setdefault("issues", []).extend(html_issues)
+            tech["issues"] = _sort_issues(tech["issues"])
+            extras["html_validation_pages"] = len(html_issues)
+
+    if get_bool(cfg, "enable_amp_audit", False) and tech is not None:
+        emit_progress("optional", "amp_audit", message="AMP canonical pairing audit")
+        amp = amp_audit_issues(df)
+        if amp:
+            tech.setdefault("issues", []).extend(amp)
+            tech["issues"] = _sort_issues(tech["issues"])
+            extras["amp_audit_issues"] = len(amp)
+
+    if get_bool(cfg, "enable_wayback_lookup", False) and tech is not None:
+        emit_progress("optional", "wayback", message="Wayback lookup for 404 URLs")
+        wb = wayback_issues(df)
+        if wb:
+            tech.setdefault("issues", []).extend(wb)
+            tech["issues"] = _sort_issues(tech["issues"])
+            extras["wayback_404_checked"] = len(wb)
+
+    if get_bool(cfg, "enable_axe", False) and a11y is not None:
+        render_mode = str(cfg.get("crawl_render_mode") or "static").strip().lower()
+        if render_mode == "static":
+            extras["axe_skipped"] = "enable_axe requires javascript or auto crawl rendering"
+            print(
+                "Warning: enable_axe is set but crawl_render_mode=static; axe runs only on browser-rendered pages.",
+                file=sys.stderr,
+                flush=True,
+            )
+        else:
+            emit_progress("optional", "axe", message="Collecting axe accessibility violations")
+            axe = axe_issues_from_df(df)
+            if axe:
+                a11y.setdefault("issues", []).extend(axe)
+                a11y["issues"] = _sort_issues(a11y["issues"])
+                extras["axe_violation_count"] = len(axe)
+
+    emit_progress("optional", "done", message="Optional audits complete")
+    return extras
diff --git a/src/website_profiling/reporting/site_files.py b/src/website_profiling/reporting/site_files.py
new file mode 100644
index 00000000..4570f834
--- /dev/null
+++ b/src/website_profiling/reporting/site_files.py
@@ -0,0 +1,151 @@
+"""Parse and fetch root site files: ads.txt, security.txt, RDAP org lookup."""
+from __future__ import annotations
+
+import re
+from typing import Any, Optional
+from urllib.parse import urlparse
+
+import requests
+
+_ADS_LINE_RE = re.compile(
+    r"^([a-z0-9.\-*]+)\s*,\s*([^,\s]+)\s*,\s*(DIRECT|RESELLER|BUYER)\s*$",
+    re.IGNORECASE,
+)
+
+
+def parse_ads_txt(text: str) -> dict[str, Any]:
+    """Parse ads.txt content. Returns present flags and validation metadata."""
+    out: dict[str, Any] = {
+        "ads_txt_present": False,
+        "ads_txt_valid": False,
+        "ads_txt_line_count": 0,
+        "ads_txt_issues": [],
+    }
+    if not text or not text.strip():
+        return out
+    out["ads_txt_present"] = True
+    valid_lines = 0
+    issues: list[str] = []
+    for i, raw in enumerate(text.splitlines(), start=1):
+        line = raw.strip()
+        if not line or line.startswith("#"):
+            continue
+        if _ADS_LINE_RE.match(line):
+            valid_lines += 1
+        else:
+            issues.append(f"invalid_line:{i}")
+    out["ads_txt_line_count"] = valid_lines
+    out["ads_txt_valid"] = valid_lines > 0 and not issues
+    if valid_lines == 0 and out["ads_txt_present"]:
+        issues.append("no_sellers")
+    out["ads_txt_issues"] = issues
+    return out
+
+
+def parse_security_txt(text: str) -> dict[str, Any]:
+    """Parse security.txt (RFC 9116)."""
+    out: dict[str, Any] = {
+        "security_txt_present": False,
+        "security_txt_valid": False,
+        "security_txt_contact": [],
+        "security_txt_expires": None,
+    }
+    if not text or not text.strip():
+        return out
+    out["security_txt_present"] = True
+    contacts: list[str] = []
+    expires: Optional[str] = None
+    recognized = 0
+    for raw in text.splitlines():
+        line = raw.strip()
+        if not line or line.startswith("#"):
+            continue
+        if ":" not in line:
+            continue
+        key, _, value = line.partition(":")
+        key = key.strip().lower()
+        value = value.strip()
+        if not value:
+            continue
+        if key == "contact":
+            recognized += 1
+            contacts.append(value)
+        elif key == "expires":
+            recognized += 1
+            expires = value
+    out["security_txt_contact"] = contacts
+    out["security_txt_expires"] = expires
+    out["security_txt_valid"] = recognized > 0
+    return out
+
+
+def fetch_ads_txt(session: requests.Session, origin: str, timeout: int = 8) -> dict[str, Any]:
+    """Fetch and parse ads.txt from site origin."""
+    defaults = parse_ads_txt("")
+    try:
+        r = session.get(f"{origin}/ads.txt", timeout=timeout)
+        if r.status_code == 200 and r.text:
+            return parse_ads_txt(r.text)
+    except Exception:
+        pass
+    return defaults
+
+
+def fetch_security_txt(session: requests.Session, origin: str, timeout: int = 8) -> dict[str, Any]:
+    """Fetch security.txt from .well-known then root."""
+    defaults = parse_security_txt("")
+    paths = ("/.well-known/security.txt", "/security.txt")
+    for path in paths:
+        try:
+            r = session.get(f"{origin}{path}", timeout=timeout)
+            if r.status_code == 200 and r.text:
+                return parse_security_txt(r.text)
+        except Exception:
+            continue
+    return defaults
+
+
+def fetch_rdap_org_name(domain: str, timeout: float = 8.0) -> Optional[str]:
+    """Best-effort RDAP lookup for registrant organization name."""
+    apex = (domain or "").strip().lower()
+    if apex.startswith("www."):
+        apex = apex[4:]
+    if not apex or "." not in apex:
+        return None
+    try:
+        r = requests.get(
+            f"https://rdap.org/domain/{apex}",
+            timeout=timeout,
+            headers={"Accept": "application/rdap+json"},
+        )
+        if r.status_code != 200:
+            return None
+        data = r.json()
+    except Exception:
+        return None
+    entities = data.get("entities") if isinstance(data, dict) else None
+    if not isinstance(entities, list):
+        return None
+    for ent in entities:
+        if not isinstance(ent, dict):
+            continue
+        roles = ent.get("roles") or []
+        if isinstance(roles, list) and "registrant" not in [str(x).lower() for x in roles]:
+            continue
+        vcard = ent.get("vcardArray")
+        if not isinstance(vcard, list) or len(vcard) < 2:
+            continue
+        for row in vcard[1:]:
+            if not isinstance(row, list) or len(row) < 4:
+                continue
+            if str(row[0]).lower() == "fn" or str(row[0]).lower() == "org":
+                val = row[3]
+                if isinstance(val, str) and val.strip():
+                    return val.strip()
+    return None
+
+
+def merge_site_file_fields(out: dict[str, Any], fields: dict[str, Any]) -> None:
+    """Merge parse/fetch result dict into site_level out."""
+    for key, val in fields.items():
+        out[key] = val
diff --git a/src/website_profiling/reporting/subdomains.py b/src/website_profiling/reporting/subdomains.py
new file mode 100644
index 00000000..73dc734d
--- /dev/null
+++ b/src/website_profiling/reporting/subdomains.py
@@ -0,0 +1,173 @@
+"""Passive subdomain inventory from crawl, GSC, and certificate transparency."""
+from __future__ import annotations
+
+from datetime import datetime, timezone
+from typing import Any
+from urllib.parse import urlparse
+
+import pandas as pd
+import requests
+
+from ..config import get_bool
+
+
+def _strip_www(host: str) -> str:
+    h = (host or "").strip().lower()
+    return h[4:] if h.startswith("www.") else h
+
+
+def _apex_from_start_url(start_url: str) -> str:
+    parsed = urlparse((start_url or "").strip())
+    return _strip_www(parsed.netloc or "")
+
+
+def _host_in_scope(host: str, apex: str) -> bool:
+    h = _strip_www(host)
+    if not h or not apex:
+        return False
+    return h == apex or h.endswith(f".{apex}")
+
+
+def _crawl_hosts(df: pd.DataFrame) -> dict[str, int]:
+    if df is None or df.empty or "url" not in df.columns:
+        return {}
+    ok = df[df["status"].astype(str).str.match(r"2\d{2}", na=False)] if "status" in df.columns else df
+    counts: dict[str, int] = {}
+    for url in ok["url"].dropna().astype(str):
+        host = (urlparse(url.strip()).netloc or "").lower()
+        if host:
+            counts[host] = counts.get(host, 0) + 1
+    return counts
+
+
+def _gsc_hosts(indexation_cov: dict[str, Any] | None) -> tuple[dict[str, int], list[str]]:
+    """Return host URL counts from GSC pages and gsc_not_crawled list."""
+    counts: dict[str, int] = {}
+    not_crawled_hosts: set[str] = set()
+    if not indexation_cov:
+        return counts, []
+    lists = indexation_cov.get("lists") if isinstance(indexation_cov.get("lists"), dict) else {}
+    for url in lists.get("gsc_not_crawled") or []:
+        host = (urlparse(str(url).strip()).netloc or "").lower()
+        if host:
+            not_crawled_hosts.add(host)
+            counts[host] = counts.get(host, 0) + 1
+    url_join = indexation_cov.get("url_join")
+    if isinstance(url_join, dict):
+        for cat in ("gsc_only",):
+            rows = url_join.get(cat)
+            if isinstance(rows, list):
+                for row in rows:
+                    if isinstance(row, dict):
+                        u = str(row.get("url") or row.get("page") or "").strip()
+                    else:
+                        u = str(row).strip()
+                    if not u:
+                        continue
+                    host = (urlparse(u).netloc or "").lower()
+                    if host:
+                        counts[host] = counts.get(host, 0) + 1
+    return counts, sorted(not_crawled_hosts)
+
+
+def _fetch_crtsh_hosts(apex: str, timeout: float = 8.0) -> tuple[set[str], str | None]:
+    """Query crt.sh for subdomains. Returns hosts and optional error message."""
+    if not apex:
+        return set(), None
+    try:
+        r = requests.get(
+            "https://crt.sh/",
+            params={"q": f"%.{apex}", "output": "json"},
+            timeout=timeout,
+            headers={"User-Agent": "WebsiteProfiling/1.0"},
+        )
+        if r.status_code != 200:
+            return set(), f"crtsh: HTTP {r.status_code}"
+        data = r.json()
+        if not isinstance(data, list):
+            return set(), "crtsh: unexpected response"
+        hosts: set[str] = set()
+        for row in data:
+            if not isinstance(row, dict):
+                continue
+            name = str(row.get("name_value") or "").strip().lower()
+            for part in name.split("\n"):
+                part = part.strip().lstrip("*.")
+                if part and "." in part:
+                    hosts.add(part)
+        return hosts, None
+    except Exception as e:
+        return set(), f"crtsh: {e}"
+
+
+def build_subdomain_inventory(
+    df: pd.DataFrame,
+    indexation_cov: dict[str, Any] | None,
+    start_url: str,
+    config: dict[str, str] | None = None,
+) -> dict[str, Any]:
+    """Build passive subdomain inventory tied to crawl and GSC coverage."""
+    if not get_bool(config or {}, "enable_subdomain_discovery", True):
+        return {"disabled": True, "apex": _apex_from_start_url(start_url), "hosts": []}
+
+    apex = _apex_from_start_url(start_url)
+    crawl_counts = _crawl_hosts(df)
+    gsc_counts, gsc_not_crawled_from_lists = _gsc_hosts(indexation_cov)
+
+    host_meta: dict[str, dict[str, Any]] = {}
+
+    def _ensure(host: str) -> dict[str, Any]:
+        h = host.lower()
+        if h not in host_meta:
+            host_meta[h] = {
+                "host": h,
+                "sources": [],
+                "in_crawl": False,
+                "in_gsc": False,
+                "url_count_crawl": 0,
+                "url_count_gsc": 0,
+                "in_scope": _host_in_scope(h, apex),
+            }
+        return host_meta[h]
+
+    for host, count in crawl_counts.items():
+        meta = _ensure(host)
+        if "crawl" not in meta["sources"]:
+            meta["sources"].append("crawl")
+        meta["in_crawl"] = True
+        meta["url_count_crawl"] = count
+
+    for host, count in gsc_counts.items():
+        meta = _ensure(host)
+        if "gsc" not in meta["sources"]:
+            meta["sources"].append("gsc")
+        meta["in_gsc"] = True
+        meta["url_count_gsc"] = max(meta["url_count_gsc"], count)
+
+    crtsh_error: str | None = None
+    if get_bool(config or {}, "subdomain_ct_lookup", True) and apex:
+        ct_hosts, crtsh_error = _fetch_crtsh_hosts(apex)
+        for host in ct_hosts:
+            meta = _ensure(host)
+            if "crtsh" not in meta["sources"]:
+                meta["sources"].append("crtsh")
+
+    gsc_hosts_not_crawled: list[str] = []
+    for host, meta in host_meta.items():
+        if meta["in_gsc"] and not meta["in_crawl"] and meta["in_scope"]:
+            gsc_hosts_not_crawled.append(host)
+    gsc_hosts_not_crawled = sorted(set(gsc_hosts_not_crawled) | set(gsc_not_crawled_from_lists))
+
+    out_of_scope: list[str] = sorted(h for h, m in host_meta.items() if not m["in_scope"])
+
+    hosts = sorted(host_meta.values(), key=lambda x: x["host"])
+    result: dict[str, Any] = {
+        "apex": apex,
+        "hosts": hosts,
+        "gsc_hosts_not_crawled": gsc_hosts_not_crawled,
+        "out_of_scope_discovered": out_of_scope,
+        "fetched_at": datetime.now(timezone.utc).isoformat(),
+    }
+    if crtsh_error:
+        result["crtsh_error"] = crtsh_error
+    return result
diff --git a/src/website_profiling/tools/audit_tools/crawl.py b/src/website_profiling/tools/audit_tools/crawl.py
index 43fb33ab..66d924ca 100644
--- a/src/website_profiling/tools/audit_tools/crawl.py
+++ b/src/website_profiling/tools/audit_tools/crawl.py
@@ -265,6 +265,7 @@ def search_pages_advanced(conn: Connection, ctx: AuditToolContext, args: dict[st
     missing_title = args.get("missing_title")
     min_word = args.get("min_word_count")
     max_word = args.get("max_word_count")
+    has_pagination = args.get("has_pagination")
 
     def _truthy(val: Any) -> bool:
         return str(val or "").lower() in ("true", "1", "yes")
@@ -301,6 +302,11 @@ def _flag(val: Any) -> bool:
                     continue
             except (TypeError, ValueError):
                 pass
+        if _flag(has_pagination):
+            pa = _parse_page_analysis(r)
+            pag = pa.get("pagination") if isinstance(pa.get("pagination"), dict) else {}
+            if not (pag.get("rel_next") or pag.get("rel_prev")):
+                continue
         filtered.append({
             "url": str(r.get("url") or ""),
             "status": str(r.get("status") or ""),
diff --git a/src/website_profiling/tools/audit_tools/export_extras.py b/src/website_profiling/tools/audit_tools/export_extras.py
new file mode 100644
index 00000000..093350f5
--- /dev/null
+++ b/src/website_profiling/tools/audit_tools/export_extras.py
@@ -0,0 +1,46 @@
+"""Extra export and validation MCP tools."""
+from __future__ import annotations
+
+from typing import Any
+
+from psycopg import Connection
+
+from ...integrations.google.rich_results import validate_urls
+from ...tools.export_sitemap import build_sitemap_xml
+from ..export_artifacts import save_artifact
+from .context import AuditToolContext
+
+
+def export_sitemap_xml(conn: Connection, ctx: AuditToolContext, args: dict[str, Any]) -> dict[str, Any]:
+    payload = ctx.load_payload(conn)
+    if not payload:
+        return {"error": "report not found"}
+    xml = build_sitemap_xml(payload)
+    artifact = save_artifact(
+        xml,
+        filename="sitemap.xml",
+        mime_type="application/xml",
+    )
+    return {**artifact, "url_count": xml.count("<loc>")}
+
+
+def validate_rich_results(conn: Connection, ctx: AuditToolContext, args: dict[str, Any]) -> dict[str, Any]:
+    payload = ctx.load_payload(conn)
+    if not payload:
+        return {"error": "report not found"}
+    link_rows = [l for l in (payload.get("links") or []) if isinstance(l, dict)]
+    urls = [str(l.get("url") or "") for l in link_rows]
+    urls = [u for u in urls if u][: int(args.get("limit") or 50)]
+    links_by_url = {str(l.get("url") or ""): l for l in link_rows if l.get("url")}
+    creds = None
+    site_url = str(args.get("site_url") or args.get("gsc_site_url") or "").strip() or None
+    if ctx.property_id:
+        try:
+            from ...integrations.google.auth import build_credentials
+
+            creds = build_credentials(ctx.property_id)
+        except Exception:
+            creds = None
+    rows = validate_urls(urls, creds=creds, site_url=site_url, links_by_url=links_by_url)
+    provenance = rows[0]["provenance"] if rows else "Estimated"
+    return {"rows": rows, "count": len(rows), "provenance": provenance}
diff --git a/src/website_profiling/tools/audit_tools/links.py b/src/website_profiling/tools/audit_tools/links.py
index b0bbfb35..f131d0fa 100644
--- a/src/website_profiling/tools/audit_tools/links.py
+++ b/src/website_profiling/tools/audit_tools/links.py
@@ -117,3 +117,50 @@ def list_broken_link_sources(conn: Connection, ctx: AuditToolContext, args: dict
     items.sort(key=lambda x: x["source_count"], reverse=True)
     sliced = cap_list(items, limit, max_cap=50)
     return {"sources": sliced["items"], "total": sliced["total"], "truncated": sliced["truncated"]}
+
+
+def get_link_rel_summary(conn: Connection, ctx: AuditToolContext, args: dict[str, Any]) -> dict[str, Any]:
+    scoped = ctx.with_args(args)
+    payload = scoped.load_payload(conn)
+    if not payload:
+        return {"error": "no report found"}
+    summary = payload.get("link_rel_summary")
+    if isinstance(summary, dict):
+        return summary
+    from ...reporting.link_edges_report import summarize_link_rel
+
+    edges = payload.get("link_edges") or []
+    return summarize_link_rel(edges if isinstance(edges, list) else [])
+
+
+def get_inlink_anchors(conn: Connection, ctx: AuditToolContext, args: dict[str, Any]) -> dict[str, Any]:
+    scoped = ctx.with_args(args)
+    payload = scoped.load_payload(conn)
+    if not payload:
+        return {"error": "no report found", "rows": [], "total": 0, "truncated": False}
+    limit = parse_limit(args.get("limit"), 50, 200)
+    target = str(args.get("url") or args.get("target_url") or "").strip().rstrip("/").lower()
+    rows = payload.get("inlink_anchor_matrix") or []
+    if not isinstance(rows, list):
+        rows = []
+    if target:
+        rows = [r for r in rows if str(r.get("target_url") or "").lower().rstrip("/") == target]
+    sliced = cap_list(rows, limit, max_cap=200)
+    return {"rows": sliced["items"], "total": sliced["total"], "truncated": sliced["truncated"]}
+
+
+def list_nofollow_internal_links(conn: Connection, ctx: AuditToolContext, args: dict[str, Any]) -> dict[str, Any]:
+    scoped = ctx.with_args(args)
+    payload = scoped.load_payload(conn)
+    if not payload:
+        return {"error": "no report found", "links": [], "total": 0, "truncated": False}
+    limit = parse_limit(args.get("limit"), 50, 100)
+    edges = payload.get("link_edges") or []
+    items = [
+        e for e in (edges if isinstance(edges, list) else [])
+        if isinstance(e, dict)
+        and e.get("link_type") == "internal"
+        and e.get("is_nofollow")
+    ]
+    sliced = cap_list(items, limit, max_cap=100)
+    return {"links": sliced["items"], "total": sliced["total"], "truncated": sliced["truncated"]}
diff --git a/src/website_profiling/tools/audit_tools/llm_tools.py b/src/website_profiling/tools/audit_tools/llm_tools.py
index 7e39bf91..a635eaab 100644
--- a/src/website_profiling/tools/audit_tools/llm_tools.py
+++ b/src/website_profiling/tools/audit_tools/llm_tools.py
@@ -95,7 +95,13 @@ def get_portfolio_summary(conn: Connection, _ctx: AuditToolContext, args: dict[s
             "generated_at": generated_at.isoformat() if hasattr(generated_at, "isoformat") else str(generated_at or ""),
             "issue_counts": issue_counts,
         })
-    return {"properties": summaries, "count": len(summaries)}
+    scores = [s["health_score"] for s in summaries if isinstance(s.get("health_score"), (int, float))]
+    median = None
+    if scores:
+        ordered = sorted(scores)
+        mid = len(ordered) // 2
+        median = ordered[mid] if len(ordered) % 2 else (ordered[mid - 1] + ordered[mid]) / 2
+    return {"properties": summaries, "count": len(summaries), "median_health_score": median}
 
 
 def expand_keywords(conn: Connection, ctx: AuditToolContext, args: dict[str, Any]) -> dict[str, Any]:
diff --git a/src/website_profiling/tools/audit_tools/property_profile.py b/src/website_profiling/tools/audit_tools/property_profile.py
new file mode 100644
index 00000000..4cdce4ba
--- /dev/null
+++ b/src/website_profiling/tools/audit_tools/property_profile.py
@@ -0,0 +1,100 @@
+"""MCP/chat tools for property profile: site files, subdomains, contacts."""
+from __future__ import annotations
+
+from typing import Any
+
+from psycopg import Connection
+
+from ._slice import cap_list, parse_limit
+from .context import AuditToolContext
+
+
+def _site_level_or_error(payload: dict[str, Any]) -> tuple[dict[str, Any] | None, dict[str, Any] | None]:
+    site_level = payload.get("site_level")
+    if not isinstance(site_level, dict):
+        return None, {"error": "site_level not in payload", "missing": True}
+    return site_level, None
+
+
+def get_ads_txt_status(conn: Connection, ctx: AuditToolContext, args: dict[str, Any]) -> dict[str, Any]:
+    scoped = ctx.with_args(args)
+    payload = scoped.load_payload(conn)
+    if not payload:
+        return {"error": "no report found"}
+    site_level, err = _site_level_or_error(payload)
+    if err:
+        return err
+    return {
+        "ads_txt_present": site_level.get("ads_txt_present"),
+        "ads_txt_valid": site_level.get("ads_txt_valid"),
+        "ads_txt_line_count": site_level.get("ads_txt_line_count"),
+        "ads_txt_issues": site_level.get("ads_txt_issues") or [],
+    }
+
+
+def get_security_txt_status(conn: Connection, ctx: AuditToolContext, args: dict[str, Any]) -> dict[str, Any]:
+    scoped = ctx.with_args(args)
+    payload = scoped.load_payload(conn)
+    if not payload:
+        return {"error": "no report found"}
+    site_level, err = _site_level_or_error(payload)
+    if err:
+        return err
+    return {
+        "security_txt_present": site_level.get("security_txt_present"),
+        "security_txt_valid": site_level.get("security_txt_valid"),
+        "security_txt_contact": site_level.get("security_txt_contact") or [],
+        "security_txt_expires": site_level.get("security_txt_expires"),
+    }
+
+
+def list_subdomains(conn: Connection, ctx: AuditToolContext, args: dict[str, Any]) -> dict[str, Any]:
+    scoped = ctx.with_args(args)
+    payload = scoped.load_payload(conn)
+    if not payload:
+        return {"error": "no report found"}
+    data = payload.get("subdomains")
+    if not isinstance(data, dict):
+        return {"error": "subdomains not in report payload", "missing": True}
+    in_scope_only = args.get("in_scope_only", True)
+    if isinstance(in_scope_only, str):
+        in_scope_only = in_scope_only.lower() not in ("false", "0", "no")
+    hosts = data.get("hosts") or []
+    if in_scope_only:
+        hosts = [h for h in hosts if isinstance(h, dict) and h.get("in_scope")]
+    limit = parse_limit(args.get("limit"), 50, 200)
+    sliced = cap_list(hosts, limit, max_cap=200)
+    return {
+        "apex": data.get("apex"),
+        "hosts": sliced["items"],
+        "total": sliced["total"],
+        "truncated": sliced["truncated"],
+        "gsc_hosts_not_crawled": data.get("gsc_hosts_not_crawled") or [],
+        "out_of_scope_discovered": data.get("out_of_scope_discovered") or [],
+    }
+
+
+def get_contact_intelligence(conn: Connection, ctx: AuditToolContext, args: dict[str, Any]) -> dict[str, Any]:
+    scoped = ctx.with_args(args)
+    payload = scoped.load_payload(conn)
+    if not payload:
+        return {"error": "no report found"}
+    data = payload.get("contact_intelligence")
+    if not isinstance(data, dict):
+        return {"error": "contact_intelligence not in report payload", "missing": True}
+    limit = parse_limit(args.get("limit"), 50, 100)
+
+    def _cap(key: str) -> list[Any]:
+        items = data.get(key) or []
+        if not isinstance(items, list):
+            return []
+        return items[:limit]
+
+    return {
+        "emails": _cap("emails"),
+        "phones": _cap("phones"),
+        "addresses": _cap("addresses"),
+        "organization_names": _cap("organization_names"),
+        "primary_contact_page": data.get("primary_contact_page"),
+        "consistency_notes": data.get("consistency_notes") or [],
+    }
diff --git a/src/website_profiling/tools/audit_tools/registry.py b/src/website_profiling/tools/audit_tools/registry.py
index a291e761..07a6ce7c 100644
--- a/src/website_profiling/tools/audit_tools/registry.py
+++ b/src/website_profiling/tools/audit_tools/registry.py
@@ -129,6 +129,7 @@
     export_list_as_csv,
     list_export_formats,
 )
+from .export_extras import export_sitemap_xml, validate_rich_results
 from .image_tools import (
     get_image_audit_summary,
     list_images_needing_attention,
@@ -157,11 +158,14 @@
     list_seo_onpage_issues,
 )
 from .links import (
+    get_inlink_anchors,
     get_link_graph_summary,
+    get_link_rel_summary,
     get_outbound_link_domains,
     get_top_linked_pages,
     get_url_fingerprints,
     list_broken_link_sources,
+    list_nofollow_internal_links,
     list_orphan_pages,
 )
 from .ops import (
@@ -178,6 +182,12 @@
     list_log_uploads,
 )
 from .properties import get_property, list_properties
+from .property_profile import (
+    get_ads_txt_status,
+    get_contact_intelligence,
+    get_security_txt_status,
+    list_subdomains,
+)
 from .report import (
     get_category_scores,
     get_critical_issues,
@@ -279,6 +289,10 @@
     "get_audit_recommendations": get_audit_recommendations,
     "get_ml_errors": get_ml_errors,
     "get_ssl_expiry_info": get_ssl_expiry_info,
+    "get_ads_txt_status": get_ads_txt_status,
+    "get_security_txt_status": get_security_txt_status,
+    "list_subdomains": list_subdomains,
+    "get_contact_intelligence": get_contact_intelligence,
     "list_audit_categories": list_audit_categories,
     "get_category_recommendations": get_category_recommendations,
     "list_issues_with_ai_fixes": list_issues_with_ai_fixes,
@@ -361,6 +375,9 @@
     "get_security_findings_summary": get_security_findings_summary,
     "list_security_findings_by_type": list_security_findings_by_type,
     "list_broken_link_sources": list_broken_link_sources,
+    "get_link_rel_summary": get_link_rel_summary,
+    "get_inlink_anchors": get_inlink_anchors,
+    "list_nofollow_internal_links": list_nofollow_internal_links,
     "search_issues": search_issues,
     "generate_content_brief": generate_content_brief,
     "get_page_coach": get_page_coach,
@@ -372,6 +389,8 @@
     "compose_custom_report": compose_custom_report,
     "export_custom_report": export_custom_report,
     "list_export_formats": list_export_formats,
+    "export_sitemap_xml": export_sitemap_xml,
+    "validate_rich_results": validate_rich_results,
     "get_image_audit_summary": get_image_audit_summary,
     "list_pages_without_lazy_images": list_pages_without_lazy_images,
     "list_pages_with_images_missing_dimensions": list_pages_with_images_missing_dimensions,
diff --git a/src/website_profiling/tools/audit_tools/report.py b/src/website_profiling/tools/audit_tools/report.py
index 9db5a0ba..20d908f2 100644
--- a/src/website_profiling/tools/audit_tools/report.py
+++ b/src/website_profiling/tools/audit_tools/report.py
@@ -125,6 +125,12 @@ def list_issues(conn: Connection, ctx: AuditToolContext, args: dict[str, Any]) -
     if url_contains:
         issues = [i for i in issues if url_contains in str(i.get("url") or "").lower()]
 
+    sort_mode = str(args.get("sort") or "").strip().lower()
+    if sort_mode == "impact":
+        from ...reporting.issue_impact import sort_issues_by_impact
+
+        issues = sort_issues_by_impact(issues)
+
     total = len(issues)
     truncated = total > limit
     return {
diff --git a/src/website_profiling/tools/audit_tools/tool_catalog.py b/src/website_profiling/tools/audit_tools/tool_catalog.py
index e0569afe..52b1abec 100644
--- a/src/website_profiling/tools/audit_tools/tool_catalog.py
+++ b/src/website_profiling/tools/audit_tools/tool_catalog.py
@@ -43,6 +43,7 @@ def _tool(name: str, description: str, properties: dict[str, Any], required: lis
             "priority": {"type": "string", "enum": ["Critical", "High", "Medium", "Low"]},
             "category_id": {"type": "string"},
             "url_contains": {"type": "string"},
+            "sort": {"type": "string", "enum": ["impact"]},
             "limit": _LIMIT,
         },
     ),
@@ -107,6 +108,23 @@ def _tool(name: str, description: str, properties: dict[str, Any], required: lis
     _tool("get_audit_recommendations", "Actionable SEO recommendation bullets from the audit.", {"property_id": _PID, "report_id": _RID}),
     _tool("get_ml_errors", "ML analysis errors (duplicates, NER, clusters) if any failed.", {"property_id": _PID, "report_id": _RID}),
     _tool("get_ssl_expiry_info", "Site TLS certificate expiry from the audit.", {"property_id": _PID, "report_id": _RID}),
+    _tool("get_ads_txt_status", "ads.txt presence and validation from site-level checks.", {"property_id": _PID, "report_id": _RID}),
+    _tool("get_security_txt_status", "security.txt presence and Contact/Expires fields from site-level checks.", {"property_id": _PID, "report_id": _RID}),
+    _tool(
+        "list_subdomains",
+        "Passive subdomain inventory from crawl, GSC, and certificate transparency.",
+        {
+            "property_id": _PID,
+            "report_id": _RID,
+            "in_scope_only": {"type": "boolean", "description": "Only hosts on the property apex (default true)"},
+            "limit": {"type": "integer", "maximum": 200},
+        },
+    ),
+    _tool(
+        "get_contact_intelligence",
+        "Business contact signals from crawl, schema, security.txt, and RDAP org (sourced).",
+        {"property_id": _PID, "report_id": _RID, "limit": {"type": "integer", "maximum": 100}},
+    ),
     _tool("list_audit_categories", "All audit categories with scores and issue counts.", {"property_id": _PID, "report_id": _RID}),
     _tool("get_category_recommendations", "Category-level recommendations for one category_id.", {"category_id": {"type": "string"}, "property_id": _PID, "report_id": _RID}, ["category_id"]),
     _tool("list_issues_with_ai_fixes", "Issues that include LLM-generated fix suggestions.", {"property_id": _PID, "report_id": _RID, "limit": _LIMIT}),
@@ -213,7 +231,10 @@ def _tool(name: str, description: str, properties: dict[str, Any], required: lis
     _tool("get_security_findings_summary", "Security findings grouped by finding_type.", {"property_id": _PID, "report_id": _RID}),
     _tool("list_security_findings_by_type", "Security findings filtered by finding_type.", {"property_id": _PID, "report_id": _RID, "finding_type": {"type": "string"}, "limit": _LIMIT}, ["finding_type"]),
     _tool("list_broken_link_sources", "Pages linking to broken URLs.", {"property_id": _PID, "report_id": _RID, "limit": _LIMIT}),
-    _tool("search_issues", "Search audit issues by message, URL, category, or priority.", {"property_id": _PID, "report_id": _RID, "message_contains": {"type": "string"}, "url_contains": {"type": "string"}, "category_id": {"type": "string"}, "priority": {"type": "string"}, "limit": _LIMIT}),
+    _tool("get_link_rel_summary", "nofollow/sponsored/UGC counts from crawl link_edges.", {"property_id": _PID, "report_id": _RID}),
+    _tool("get_inlink_anchors", "Inlink anchor text matrix (target × anchor counts).", {"property_id": _PID, "report_id": _RID, "url": _URL, "limit": {"type": "integer", "maximum": 200}}),
+    _tool("list_nofollow_internal_links", "Internal links with rel=nofollow from crawl.", {"property_id": _PID, "report_id": _RID, "limit": _LIMIT}),
+    _tool("search_issues", "Search audit issues by message, URL, category, or priority.", {"property_id": _PID, "report_id": _RID, "message_contains": {"type": "string"}, "url_contains": {"type": "string"}, "category_id": {"type": "string"}, "priority": {"type": "string"}, "sort": {"type": "string", "enum": ["impact"]}, "limit": _LIMIT}),
     # LLM / portfolio
     _tool("generate_content_brief", "Generate a content brief for a target keyword.", {"property_id": _PID, "keyword": {"type": "string"}, "gaps": {"type": "array", "items": {"type": "string"}}}, ["keyword"]),
     _tool("get_page_coach", "LLM internal linking coach for one URL.", {"url": _URL, "property_id": _PID, "report_id": _RID, "refresh": {"type": "boolean"}}, ["url"]),
@@ -244,6 +265,16 @@ def _tool(name: str, description: str, properties: dict[str, Any], required: lis
         },
         ["tool_name"],
     ),
+    _tool(
+        "export_sitemap_xml",
+        "Generate XML sitemap from indexable crawled URLs.",
+        {"property_id": _PID, "report_id": _RID},
+    ),
+    _tool(
+        "validate_rich_results",
+        "Validate structured data / Rich Results for sample URLs (Estimated without API key).",
+        {"property_id": _PID, "report_id": _RID, "limit": {"type": "integer", "maximum": 50}, "api_key": {"type": "string"}},
+    ),
     _tool(
         "compose_custom_report",
         "Compose a multi-section custom report spec (tool, executive_summary, category_scores, notes sections).",
diff --git a/src/website_profiling/tools/export_crawl_workbook.py b/src/website_profiling/tools/export_crawl_workbook.py
new file mode 100644
index 00000000..87bbdb2d
--- /dev/null
+++ b/src/website_profiling/tools/export_crawl_workbook.py
@@ -0,0 +1,106 @@
+"""Export crawl workbook as ZIP of CSV sheets."""
+from __future__ import annotations
+
+import csv
+import io
+import json
+import zipfile
+from typing import Any
+
+
+def _csv_bytes(rows: list[dict[str, Any]], columns: list[str]) -> bytes:
+    buf = io.StringIO()
+    writer = csv.DictWriter(buf, fieldnames=columns, extrasaction="ignore")
+    writer.writeheader()
+    for row in rows:
+        writer.writerow({k: row.get(k, "") for k in columns})
+    return buf.getvalue().encode("utf-8")
+
+
+def _parse_custom_fields(raw: Any) -> dict[str, str]:
+    if raw is None:
+        return {}
+    if isinstance(raw, dict):
+        return {str(k): str(v) for k, v in raw.items()}
+    text = str(raw).strip()
+    if not text:
+        return {}
+    try:
+        parsed = json.loads(text)
+    except json.JSONDecodeError:
+        return {}
+    if not isinstance(parsed, dict):
+        return {}
+    return {str(k): str(v) for k, v in parsed.items()}
+
+
+def _custom_field_rows(links: list[Any]) -> tuple[list[dict[str, Any]], list[str]]:
+    rows: list[dict[str, Any]] = []
+    field_names: set[str] = set()
+    for row in links:
+        if not isinstance(row, dict):
+            continue
+        url = row.get("url")
+        custom_extract = row.get("custom_extract")
+        fields = _parse_custom_fields(row.get("custom_fields"))
+        if not url or (not custom_extract and not fields):
+            continue
+        field_names.update(fields.keys())
+        rows.append({"url": url, "custom_extract": custom_extract or "", **fields})
+    columns = ["url", "custom_extract", *sorted(field_names)]
+    return rows, columns
+
+
+def build_crawl_workbook_zip(report_payload: dict[str, Any]) -> bytes:
+    """Build ZIP containing Internal URLs, Links, Redirects, Issues CSVs."""
+    mem = io.BytesIO()
+    with zipfile.ZipFile(mem, "w", zipfile.ZIP_DEFLATED) as zf:
+        links = report_payload.get("links") or []
+        if isinstance(links, list) and links:
+            url_cols = [
+                "url", "status", "title", "meta_description", "h1",
+                "canonical_url", "inlinks", "outlinks", "depth", "word_count",
+            ]
+            zf.writestr("internal_urls.csv", _csv_bytes(links, url_cols))
+
+        link_edges = report_payload.get("link_edges") or []
+        if isinstance(link_edges, list) and link_edges:
+            edge_cols = [
+                "from_url", "to_url", "anchor_text", "rel",
+                "is_nofollow", "is_sponsored", "is_ugc", "link_type",
+            ]
+            zf.writestr("links.csv", _csv_bytes(link_edges, edge_cols))
+
+        redirects = report_payload.get("redirects") or []
+        if isinstance(redirects, list) and redirects:
+            zf.writestr(
+                "redirects.csv",
+                _csv_bytes(redirects, ["url", "message", "priority", "recommendation"]),
+            )
+
+        issue_rows: list[dict[str, Any]] = []
+        for cat in report_payload.get("categories") or []:
+            if not isinstance(cat, dict):
+                continue
+            cat_name = cat.get("name") or cat.get("id") or ""
+            for iss in cat.get("issues") or []:
+                if isinstance(iss, dict):
+                    issue_rows.append({**iss, "category": cat_name})
+        if issue_rows:
+            zf.writestr(
+                "issues.csv",
+                _csv_bytes(
+                    issue_rows,
+                    [
+                        "category", "priority", "message", "url",
+                        "impact_score", "gsc_clicks", "gsc_impressions", "ga4_sessions",
+                        "recommendation",
+                    ],
+                ),
+            )
+
+        custom_rows, custom_cols = _custom_field_rows(links if isinstance(links, list) else [])
+        if custom_rows:
+            zf.writestr("custom_fields.csv", _csv_bytes(custom_rows, custom_cols))
+
+    return mem.getvalue()
diff --git a/src/website_profiling/tools/export_sitemap.py b/src/website_profiling/tools/export_sitemap.py
new file mode 100644
index 00000000..7f912211
--- /dev/null
+++ b/src/website_profiling/tools/export_sitemap.py
@@ -0,0 +1,28 @@
+"""Generate XML sitemap from crawled indexable URLs."""
+from __future__ import annotations
+
+from typing import Any
+from xml.sax.saxutils import escape
+
+
+def build_sitemap_xml(report_payload: dict[str, Any], *, max_urls: int = 50000) -> str:
+    urls: list[str] = []
+    for row in report_payload.get("links") or []:
+        if not isinstance(row, dict):
+            continue
+        if row.get("noindex"):
+            continue
+        status = str(row.get("status") or "")
+        if not status.startswith("2"):
+            continue
+        u = str(row.get("url") or "").strip()
+        if u:
+            urls.append(u)
+    urls = urls[: max(1, int(max_urls))]
+    body = "\n".join(f"  <url><loc>{escape(u)}</loc></url>" for u in urls)
+    return (
+        '<?xml version="1.0" encoding="UTF-8"?>\n'
+        '<urlset xmlns="http://www.sitemaps.org/schemas/sitemap/0.9">\n'
+        f"{body}\n"
+        "</urlset>\n"
+    )
diff --git a/src/website_profiling/tools/plot.py b/src/website_profiling/tools/plot.py
index efdd30a0..3a548370 100644
--- a/src/website_profiling/tools/plot.py
+++ b/src/website_profiling/tools/plot.py
@@ -29,7 +29,10 @@ def run_plot(
     if not use_database:
         raise ValueError("Plot requires DATABASE_URL (PostgreSQL).")
 
+    from ..progress import emit_progress
+
     run_id = None
+    emit_progress("plot", "load_data", message="Loading crawl and edges from database")
     print("  Loading crawl and edges from DB...", flush=True)
     from ..db import db_session, get_crawl_run_info, get_latest_crawl_run_id, read_crawl, read_edges
     with db_session() as conn:
@@ -54,6 +57,7 @@ def run_plot(
     mode = (render_mode or "static").strip().lower()
 
     if not edges and not df.empty:
+        emit_progress("plot", "build_edges", message="Building link graph from crawl data")
         print("  Building edges from crawl data...", flush=True)
         edges = build_edges_from_df(
             df,
@@ -74,6 +78,7 @@ def run_plot(
 
     if edges:
         edges_df = pd.DataFrame(edges, columns=["from", "to"])
+        emit_progress("plot", "write_db", message="Writing nodes and edges to database")
         print("  Writing edges and nodes to DB...", flush=True)
         from ..db import db_session, get_latest_crawl_run_id, write_edges as db_write_edges, write_nodes as db_write_nodes
         with db_session() as conn:
diff --git a/tests/test_audit_tools_coverage.py b/tests/test_audit_tools_coverage.py
index 84922341..ec482165 100644
--- a/tests/test_audit_tools_coverage.py
+++ b/tests/test_audit_tools_coverage.py
@@ -578,6 +578,10 @@ def _read_pair(_conn: MagicMock, rid: int) -> dict:
         assert idx_mod.list_indexation_gaps(conn, ctx, {"gap_type": "sitemap_only"})["error"]
         assert idx_mod.get_indexation_url_join(conn, ctx, {})["error"]
 
+    with patch.object(Ctx, "load_payload", return_value={"indexation_coverage": "bad"}):
+        assert idx_mod.list_indexation_gaps(conn, ctx, {"gap_type": "sitemap_only"})["missing"] is True
+        assert idx_mod.get_indexation_url_join(conn, ctx, {})["missing"] is True
+
     ops_row = MagicMock()
     ops_row.keys = MagicMock(return_value=["schedule_cron", "alert_webhook_url", "alert_email"])
     ops_conn3 = MagicMock()
@@ -648,6 +652,75 @@ def _read_pair(_conn: MagicMock, rid: int) -> dict:
     assert health_mod.get_health_history(fake_h3, Ctx(property_id=1), {})["count"] == 1
 
 
+def test_property_profile_tools() -> None:
+    from website_profiling.tools.audit_tools import property_profile as pp_mod
+
+    conn = MagicMock()
+    ctx = Ctx(property_id=1)
+
+    with patch.object(Ctx, "load_payload", return_value={}):
+        assert pp_mod.get_ads_txt_status(conn, ctx, {})["error"] == "no report found"
+        assert pp_mod.get_security_txt_status(conn, ctx, {})["error"] == "no report found"
+        assert pp_mod.list_subdomains(conn, ctx, {})["error"] == "no report found"
+        assert pp_mod.get_contact_intelligence(conn, ctx, {})["error"] == "no report found"
+
+    payload = {
+        "site_level": {
+            "ads_txt_present": True,
+            "ads_txt_valid": True,
+            "ads_txt_line_count": 1,
+            "ads_txt_issues": [],
+            "security_txt_present": True,
+            "security_txt_valid": True,
+            "security_txt_contact": ["mailto:a@b.com"],
+            "security_txt_expires": "2030-01-01",
+        },
+        "subdomains": {
+            "apex": "example.com",
+            "hosts": [
+                {"host": "www.example.com", "in_scope": True},
+                {"host": "other.net", "in_scope": False},
+            ],
+            "gsc_hosts_not_crawled": ["blog.example.com"],
+            "out_of_scope_discovered": ["other.net"],
+        },
+        "contact_intelligence": {
+            "emails": [{"value": "a@b.com"}],
+            "phones": [{"value": "+1"}],
+            "addresses": [{"value": "1 Main"}],
+            "organization_names": [{"value": "Example Co"}],
+            "primary_contact_page": "https://example.com/contact",
+            "consistency_notes": ["note"],
+        },
+    }
+    with patch.object(Ctx, "load_payload", return_value=payload):
+        assert pp_mod.get_ads_txt_status(conn, ctx, {})["ads_txt_present"] is True
+        assert pp_mod.get_security_txt_status(conn, ctx, {})["security_txt_contact"]
+        scoped = pp_mod.list_subdomains(conn, ctx, {"in_scope_only": "false", "limit": 1})
+        assert scoped["truncated"] is True
+        in_scope = pp_mod.list_subdomains(conn, ctx, {"in_scope_only": True})
+        assert all(h["in_scope"] for h in in_scope["hosts"])
+        contacts = pp_mod.get_contact_intelligence(conn, ctx, {"limit": 5})
+        assert contacts["emails"][0]["value"] == "a@b.com"
+
+    with patch.object(Ctx, "load_payload", return_value={"site_level": "bad"}):
+        assert pp_mod.get_ads_txt_status(conn, ctx, {})["missing"] is True
+        assert pp_mod.get_security_txt_status(conn, ctx, {})["missing"] is True
+
+    with patch.object(Ctx, "load_payload", return_value={"subdomains": None}):
+        assert pp_mod.list_subdomains(conn, ctx, {})["missing"] is True
+
+    with patch.object(Ctx, "load_payload", return_value={"contact_intelligence": "bad"}):
+        assert pp_mod.get_contact_intelligence(conn, ctx, {})["missing"] is True
+
+    with patch.object(
+        Ctx,
+        "load_payload",
+        return_value={"contact_intelligence": {"emails": "bad", "phones": [], "addresses": [], "organization_names": []}},
+    ):
+        assert pp_mod.get_contact_intelligence(conn, ctx, {})["emails"] == []
+
+
 def test_all_tools_empty_payload() -> None:
     """Hit 'no report found' branches across payload-backed tools."""
     conn = MagicMock()
@@ -665,6 +738,7 @@ def test_all_tools_empty_payload() -> None:
         "get_crux_summary", "get_tech_stack_summary", "get_security_findings",
         "get_category_issues",
         "get_audit_recommendations", "get_ml_errors", "get_ssl_expiry_info",
+        "get_ads_txt_status", "get_security_txt_status", "list_subdomains", "get_contact_intelligence",
         "list_audit_categories", "get_category_recommendations", "list_issues_with_ai_fixes",
         "list_seo_onpage_issues", "list_content_url_issues", "list_pages_missing_title",
         "get_crawl_summary", "get_mime_type_breakdown", "list_indexation_gaps",
diff --git a/tests/test_audit_tools_expanded.py b/tests/test_audit_tools_expanded.py
index 83e51408..59fcad20 100644
--- a/tests/test_audit_tools_expanded.py
+++ b/tests/test_audit_tools_expanded.py
@@ -161,7 +161,7 @@ def conn() -> MagicMock:
 def test_handler_schema_parity() -> None:
     names = {t["name"] for t in TOOL_DEFINITIONS}
     assert names == tool_handler_names()
-    assert len(TOOL_DEFINITIONS) == 171
+    assert len(TOOL_DEFINITIONS) == 180
 
 
 def test_slice_helpers() -> None:
diff --git a/tests/test_audit_tools_links_extras.py b/tests/test_audit_tools_links_extras.py
new file mode 100644
index 00000000..2d9ee34b
--- /dev/null
+++ b/tests/test_audit_tools_links_extras.py
@@ -0,0 +1,100 @@
+"""Tests for link edge audit tools and export extras."""
+from __future__ import annotations
+
+from unittest.mock import MagicMock
+
+import pytest
+
+from website_profiling.tools.audit_tools.context import AuditToolContext
+from website_profiling.tools.audit_tools.export_extras import export_sitemap_xml, validate_rich_results
+from website_profiling.tools.audit_tools.links import (
+    get_inlink_anchors,
+    get_link_rel_summary,
+    list_broken_link_sources,
+    list_nofollow_internal_links,
+)
+
+
+@pytest.fixture
+def ctx():
+    return AuditToolContext(property_id=1)
+
+
+def test_link_tools_with_rich_edges(ctx):
+    payload = {
+        "issues": {"broken": [{"url": "https://example.com/broken"}]},
+        "graph_edges": [
+            {"from": "https://example.com/", "to": "https://example.com/broken"},
+            ["https://example.com/a", "https://example.com/broken"],
+            "skip",
+        ],
+        "link_edges": [
+            {
+                "from_url": "https://example.com/",
+                "to_url": "https://example.com/x",
+                "anchor_text": "X",
+                "link_type": "internal",
+                "is_nofollow": True,
+            }
+        ],
+        "inlink_anchor_matrix": [
+            {"target_url": "https://example.com/x", "anchor_text": "X", "inlink_count": 2},
+            {"target_url": "https://example.com/y", "anchor_text": "Y", "inlink_count": 1},
+        ],
+    }
+    conn = MagicMock()
+    scoped = MagicMock()
+    scoped.load_payload.return_value = payload
+    ctx.with_args = MagicMock(return_value=scoped)
+
+    broken = list_broken_link_sources(conn, ctx, {"limit": 10})
+    assert broken["sources"][0]["source_count"] >= 2
+
+    rel = get_link_rel_summary(conn, ctx, {})
+    assert rel["total_edges"] == 1
+
+    anchors = get_inlink_anchors(conn, ctx, {"url": "https://example.com/x"})
+    assert len(anchors["rows"]) == 1
+
+    nofollow = list_nofollow_internal_links(conn, ctx, {})
+    assert nofollow["links"][0]["is_nofollow"] is True
+
+
+def test_get_link_rel_summary_uses_payload_summary(ctx):
+    conn = MagicMock()
+    scoped = MagicMock()
+    scoped.load_payload.return_value = {"link_rel_summary": {"total_edges": 9}}
+    ctx.with_args = MagicMock(return_value=scoped)
+    assert get_link_rel_summary(conn, ctx, {})["total_edges"] == 9
+
+
+def test_export_extras_empty_payload(ctx):
+    conn = MagicMock()
+    ctx.load_payload = MagicMock(return_value={})
+    assert export_sitemap_xml(conn, ctx, {})["error"] == "report not found"
+    assert validate_rich_results(conn, ctx, {})["error"] == "report not found"
+
+
+def test_export_sitemap_xml_tool(ctx, monkeypatch):
+    conn = MagicMock()
+    scoped = MagicMock()
+    scoped.load_payload.return_value = {"links": [{"url": "https://example.com/", "status": "200"}]}
+    ctx.load_payload = scoped.load_payload
+    monkeypatch.setattr(
+        "website_profiling.tools.audit_tools.export_extras.save_artifact",
+        lambda content, **kwargs: {"path": "/tmp/sitemap.xml", "filename": kwargs.get("filename")},
+    )
+    out = export_sitemap_xml(conn, ctx, {})
+    assert out["url_count"] == 1
+
+
+def test_validate_rich_results_tool(ctx):
+    conn = MagicMock()
+    scoped = MagicMock()
+    scoped.load_payload.return_value = {
+        "links": [{"url": "https://example.com/", "status": "200", "has_schema": True, "page_analysis": {}}]
+    }
+    ctx.load_payload = scoped.load_payload
+    out = validate_rich_results(conn, ctx, {"limit": 5})
+    assert out["count"] == 1
+    assert out["rows"][0]["provenance"] == "Crawl analysis"
diff --git a/tests/test_config_schema_keys.py b/tests/test_config_schema_keys.py
index a000d69a..97c211a4 100644
--- a/tests/test_config_schema_keys.py
+++ b/tests/test_config_schema_keys.py
@@ -21,6 +21,16 @@
     "preserve_crawl_history",
     "crawl_stream_to_db",
     "crawl_exclude_urls",
+    "crawl_discovery_mode",
+    "crawl_url_list",
+    "crawl_user_agent_preset",
+    "crawl_user_agent_custom",
+    "crawl_auth_username",
+    "crawl_auth_password",
+    "crawl_extra_headers",
+    "crawl_cookies",
+    "crawl_robots_txt_override",
+    "custom_extractors",
     "crawl_render_mode",
     "crawl_js_concurrency",
     "crawl_js_timeout",
@@ -54,6 +64,13 @@
     "run_lighthouse",
     "run_lighthouse_on_pages",
     "enable_crux",
+    "enable_rich_results_validation",
+    "google_rich_results_api_key",
+    "enable_axe",
+    "enable_spell_check",
+    "enable_html_validation",
+    "enable_amp_audit",
+    "enable_wayback_lookup",
     "competitor_domains",
     "bing_webmaster_api_key",
     "serp_api_key",
@@ -75,6 +92,9 @@
     "enable_google_analytics",
     "google_date_range_days",
     "google_url_gap_list_limit",
+    "enable_subdomain_discovery",
+    "subdomain_ct_lookup",
+    "enable_rdap_org_lookup",
     "enrich_keywords_after_report",
     "keyword_max_pages",
     "keyword_gsc_max_rows",
@@ -107,6 +127,12 @@ def test_pipeline_example_keys_are_subset_of_schema():
 def test_schema_keys_documented_or_optional_in_example():
     keys = parse_config_keys(REPO_ROOT / "pipeline-config.example.txt")
     # Tristate 'auto' keys may be omitted from static examples (UI omits them when auto).
-    optional_omitted = {"enrich_keywords_after_report", "active_property_id"}
+    optional_omitted = {
+        "enrich_keywords_after_report",
+        "active_property_id",
+        "enable_subdomain_discovery",
+        "subdomain_ct_lookup",
+        "enable_rdap_org_lookup",
+    }
     missing = SCHEMA_KEYS - keys - optional_omitted
     assert not missing, f"pipeline-config.example.txt missing keys: {sorted(missing)}"
diff --git a/tests/test_contrast_issues.py b/tests/test_contrast_issues.py
new file mode 100644
index 00000000..81f08677
--- /dev/null
+++ b/tests/test_contrast_issues.py
@@ -0,0 +1,147 @@
+"""Tests for accessibility contrast issue detection."""
+from __future__ import annotations
+
+import json
+
+import pandas as pd
+
+from website_profiling.reporting.categories import (
+    category_html_accessibility,
+    contrast_issues_from_sources,
+)
+
+
+def test_contrast_from_axe_violations():
+    pa = {
+        "axe_violations": [
+            {
+                "id": "color-contrast",
+                "description": "Elements must have sufficient color contrast",
+                "help": "Fix contrast",
+            }
+        ]
+    }
+    df = pd.DataFrame([
+        {
+            "url": "https://ex.com/page",
+            "status": "200",
+            "page_analysis": json.dumps(pa),
+            "h1_count": 1,
+            "images_total": 0,
+            "images_without_alt": 0,
+        }
+    ])
+    issues = contrast_issues_from_sources(df, {})
+    assert len(issues) == 1
+    assert "axe" in issues[0]["message"].lower()
+    assert issues[0]["url"] == "https://ex.com/page"
+
+
+def test_contrast_from_lighthouse_failures():
+    df = pd.DataFrame([
+        {"url": "https://ex.com/a", "status": "200", "h1_count": 1, "images_total": 0, "images_without_alt": 0},
+    ])
+    lh = {
+        "https://ex.com/a": {
+            "url": "https://ex.com/a",
+            "top_failures": [{"id": "color-contrast", "helpText": "Background and foreground colors do not contrast"}],
+        }
+    }
+    issues = contrast_issues_from_sources(df, lh)
+    assert len(issues) == 1
+    assert "Lighthouse" in issues[0]["message"]
+
+
+def test_category_omits_stub_when_contrast_data_exists():
+    pa = {"axe_violations": [{"id": "color-contrast", "description": "bad contrast", "help": "fix"}]}
+    df = pd.DataFrame([
+        {
+            "url": "https://ex.com/",
+            "status": "200",
+            "page_analysis": json.dumps(pa),
+            "h1_count": 1,
+            "images_total": 0,
+            "images_without_alt": 0,
+            "word_count": 200,
+            "reading_level": 8,
+        }
+    ])
+    cat = category_html_accessibility(df)
+    messages = [i["message"] for i in cat["issues"]]
+    assert not any("not measured" in m for m in messages)
+    assert any("axe" in m.lower() for m in messages)
+
+
+def test_contrast_skips_non_matching_axe_and_lighthouse_rows():
+    df = pd.DataFrame([
+        {
+            "url": "",
+            "status": "200",
+            "page_analysis": json.dumps({"axe_violations": [{"id": "label", "description": "x"}]}),
+        },
+        {
+            "url": "https://ex.com/b",
+            "status": "200",
+            "page_analysis": json.dumps({"axe_violations": [{"id": "label"}]}),
+        },
+    ])
+    lh = {
+        "https://ex.com/a": "not-a-dict",
+        "https://ex.com/b": {
+            "top_failures": [
+                "bad",
+                {"id": "image-alt", "helpText": "missing alt"},
+            ],
+        },
+    }
+    assert contrast_issues_from_sources(df, lh) == []
+
+
+def test_contrast_deduplicates_lighthouse_when_axe_already_reported():
+    pa = {"axe_violations": [{"id": "color-contrast", "description": "axe contrast", "help": "fix"}]}
+    df = pd.DataFrame([
+        {"url": "https://ex.com/shared", "status": "200", "page_analysis": json.dumps(pa)},
+    ])
+    lh = {
+        "https://ex.com/shared": {
+            "top_failures": [{"id": "color-contrast", "helpText": "lh contrast"}],
+        }
+    }
+    issues = contrast_issues_from_sources(df, lh)
+    assert len(issues) == 1
+    assert "axe" in issues[0]["message"].lower()
+
+
+def test_contrast_accepts_dict_page_analysis_cell():
+    df = pd.DataFrame([
+        {
+            "url": "https://ex.com/dict",
+            "status": "200",
+            "page_analysis": {"axe_violations": [{"id": "color-contrast", "description": "low contrast"}]},
+        },
+        {
+            "url": "https://ex.com/bad-json",
+            "status": "200",
+            "page_analysis": "{not-json",
+        },
+    ])
+    issues = contrast_issues_from_sources(df, {})
+    assert len(issues) == 1
+    assert issues[0]["url"] == "https://ex.com/dict"
+
+
+def test_category_keeps_stub_without_contrast_data():
+    df = pd.DataFrame([
+        {
+            "url": "https://ex.com/",
+            "status": "200",
+            "page_analysis": "{}",
+            "h1_count": 1,
+            "images_total": 0,
+            "images_without_alt": 0,
+            "word_count": 200,
+            "reading_level": 8,
+        }
+    ])
+    cat = category_html_accessibility(df)
+    assert any("not measured" in i["message"] for i in cat["issues"])
diff --git a/tests/test_crawl_discovery.py b/tests/test_crawl_discovery.py
new file mode 100644
index 00000000..23616e9c
--- /dev/null
+++ b/tests/test_crawl_discovery.py
@@ -0,0 +1,79 @@
+"""Tests for crawl discovery modes."""
+from __future__ import annotations
+
+import json
+
+import website_profiling.crawl.crawler as mod
+from website_profiling.crawl.discovery import (
+    follow_links_for_mode,
+    normalize_discovery_mode,
+    parse_crawl_url_list,
+    seed_sitemap_for_mode,
+)
+from website_profiling.crawl.fetchers.base import FetchResult
+
+
+def test_normalize_discovery_mode_defaults_spider():
+    assert normalize_discovery_mode(None) == "spider"
+    assert normalize_discovery_mode("invalid") == "spider"
+    assert normalize_discovery_mode("LIST") == "list"
+
+
+def test_parse_crawl_url_list_dedupes():
+    raw = "https://a.com/1\nhttps://a.com/1,https://a.com/2"
+    assert parse_crawl_url_list(raw) == ["https://a.com/1", "https://a.com/2"]
+
+
+def test_mode_flags():
+    assert follow_links_for_mode("list") is False
+    assert follow_links_for_mode("sitemap") is False
+    assert follow_links_for_mode("spider") is True
+    assert seed_sitemap_for_mode("list") is False
+    assert seed_sitemap_for_mode("sitemap") is True
+
+
+def test_list_mode_does_not_follow_links(monkeypatch):
+    monkeypatch.setattr(
+        "website_profiling.crawl.sitemap.discover_sitemap_urls",
+        lambda *_a, **_k: [],
+    )
+    c = mod.Crawler(
+        start_url="https://site.com",
+        ignore_robots=True,
+        store_outlinks=True,
+        discovery_mode="list",
+        crawl_url_list=["https://site.com/a", "https://site.com/b"],
+    )
+    assert c.follow_links is False
+    assert set(c.depths.keys()) == {"https://site.com/a", "https://site.com/b"}
+
+    c.fetch = lambda url: FetchResult(  # type: ignore[method-assign]
+        status=200,
+        content_type="text/html",
+        text='<html><a href="https://site.com/c">c</a></html>',
+        response_time_ms=1,
+        content_length=10,
+        final_url=url,
+        headers_dict={},
+        redirect_chain_length=0,
+        fetch_method="static",
+    )
+    out = c.worker("https://site.com/a")
+    assert out["status"] == 200
+    assert "https://site.com/c" not in c.depths
+
+
+def test_parse_link_edges_rel_flags():
+    from website_profiling.common import parse_link_edges
+
+    html = """
+    <html><body>
+      <a href="/a" rel="nofollow sponsored">A</a>
+      <a href="/b">B</a>
+    </body></html>
+    """
+    _title, edges = parse_link_edges("https://site.com/page", html)
+    by_href = {e["to_url"].split("/")[-1]: e for e in edges}
+    assert by_href["a"]["is_nofollow"] is True
+    assert by_href["a"]["is_sponsored"] is True
+    assert by_href["b"]["is_nofollow"] is False
diff --git a/tests/test_crawl_fetchers.py b/tests/test_crawl_fetchers.py
index 15837e85..6df58015 100644
--- a/tests/test_crawl_fetchers.py
+++ b/tests/test_crawl_fetchers.py
@@ -381,6 +381,21 @@ def test_build_fetcher_unknown_mode_falls_back_to_static():
         fetcher.close()
 
 
+def test_browser_auth_from_session_maps_headers_and_credentials():
+    import requests
+
+    from website_profiling.crawl.fetchers.factory import _browser_auth_from_session
+
+    session = requests.Session()
+    session.headers.update({"User-Agent": "TestBot", "Cookie": "sid=abc", "X-Auth": "token"})
+    session.auth = ("crawler", "secret")
+    headers, credentials = _browser_auth_from_session(session)
+    assert "User-Agent" not in headers
+    assert headers["Cookie"] == "sid=abc"
+    assert headers["X-Auth"] == "token"
+    assert credentials == {"username": "crawler", "password": "secret"}
+
+
 def test_ensure_browser_deps_installs_when_missing(monkeypatch):
     pip_called: list[str] = []
     pw_called: list[str] = []
diff --git a/tests/test_crawl_gap_coverage.py b/tests/test_crawl_gap_coverage.py
new file mode 100644
index 00000000..aa14403a
--- /dev/null
+++ b/tests/test_crawl_gap_coverage.py
@@ -0,0 +1,422 @@
+"""Coverage for crawl discovery, extraction, axe, auth, and link_edges store."""
+from __future__ import annotations
+
+import asyncio
+import json
+from unittest.mock import AsyncMock, MagicMock
+
+import pytest
+
+from tests.db_test_fakes import CrawlConn
+from tests.test_browser_fetcher_unit import _install_fake_playwright
+
+
+def test_parse_crawl_url_list_resolves_relative_urls():
+    from website_profiling.crawl.discovery import parse_crawl_url_list
+
+    out = parse_crawl_url_list("/about\n, /contact", start_url="https://site.com")
+    assert out == ["https://site.com/about", "https://site.com/contact"]
+
+
+def test_parse_extractors_config_invalid_json():
+    from website_profiling.crawl.extraction import parse_extractors_config, run_extractors
+
+    assert parse_extractors_config("{bad") == []
+    assert run_extractors("", [{"name": ""}]) == {}
+
+
+def test_run_extractors_regex_css_xpath():
+    from website_profiling.crawl.extraction import run_extractors
+
+    html = """
+    <html><body>
+      <span id="sku">ABC</span>
+      <p class="price">$12</p>
+    </body></html>
+    """
+    specs = [
+        {"name": "sku", "type": "css", "selector": "#sku"},
+        {"name": "price", "type": "css", "selector": ".price"},
+        {"name": "num", "type": "regex", "pattern": r"(\d+)"},
+        {"name": "xpath_val", "type": "xpath", "expr": "//span[@id='sku']/text()"},
+    ]
+    out = run_extractors(html, specs)
+    assert out["sku"] == "ABC"
+    assert out["price"] == "$12"
+    assert out["num"] == "12"
+    assert out["xpath_val"] == "ABC"
+
+
+def test_run_extractors_skips_bad_xpath():
+    from website_profiling.crawl.extraction import run_extractors
+
+    html = "<html><body>x</body></html>"
+    out = run_extractors(html, [{"name": "bad", "type": "xpath", "expr": "//["}])
+    assert out == {}
+
+
+def test_run_extractors_skips_empty_specs():
+    from website_profiling.crawl.extraction import run_extractors
+
+    html = "<html><body><span id='x'>1</span></body></html>"
+    specs = [
+        {"name": "", "type": "css", "selector": "#x"},
+        {"name": "missing_sel", "type": "css", "selector": ""},
+        {"name": "missing_el", "type": "css", "selector": "#nope"},
+        {"name": "missing_xpath", "type": "xpath", "expr": ""},
+        {"name": "missing_regex", "type": "regex", "pattern": ""},
+    ]
+    assert run_extractors(html, specs) == {}
+
+
+def test_run_axe_on_page_returns_violations():
+    from website_profiling.crawl.axe_runner import run_axe_on_page
+
+    async def _run() -> None:
+        page = MagicMock()
+        page.add_script_tag = AsyncMock()
+        page.evaluate = AsyncMock(
+            return_value=[
+                {
+                    "id": "color-contrast",
+                    "impact": "serious",
+                    "description": "Low contrast",
+                    "help": "Fix colors",
+                    "nodes": 2,
+                }
+            ]
+        )
+        rows = await run_axe_on_page(page)
+        assert rows and rows[0]["id"] == "color-contrast"
+
+    asyncio.run(_run())
+
+
+def test_run_axe_on_page_swallows_errors():
+    from website_profiling.crawl.axe_runner import run_axe_on_page
+
+    async def _run() -> None:
+        page = MagicMock()
+        page.add_script_tag = AsyncMock(side_effect=RuntimeError("boom"))
+        assert await run_axe_on_page(page) == []
+
+    asyncio.run(_run())
+
+
+def test_parse_link_edges_image_anchor_text():
+    from website_profiling.common import parse_link_edges
+
+    html = '<html><body><a href="/x"><img src="/i.png" alt="">Click</a></body></html>'
+    _, edges = parse_link_edges("https://site.com/", html)
+    assert edges[0]["anchor_text"] == "[image] Click"
+
+
+def test_analyze_html_pagination_links():
+    from website_profiling.analysis.page import analyze_html
+
+    html = """
+    <html><head>
+      <link rel="next" href="/page/2">
+      <link rel="prev" href="/page/1">
+      <link rel="amphtml" href="https://site.com/amp/page/2">
+    </head><body></body></html>
+    """
+    out = analyze_html(html, "https://site.com/page/2", "https://site.com/page/2")
+    pag = out["pagination"]
+    assert pag["rel_next"] == "https://site.com/page/2"
+    assert pag["rel_prev"] == "https://site.com/page/1"
+    assert pag["amphtml"] == "https://site.com/amp/page/2"
+
+
+def test_crawler_auth_headers_and_custom_fields(monkeypatch) -> None:
+    from website_profiling.crawl.crawler import Crawler
+    from website_profiling.crawl.fetchers.base import FetchResult
+
+    monkeypatch.setattr(
+        "website_profiling.crawl.sitemap.discover_sitemap_urls",
+        lambda *_a, **_k: [],
+    )
+    captured: dict = {}
+
+    def fake_build_fetcher(**kwargs):
+        captured.update(kwargs)
+        fetcher = MagicMock()
+        fetcher.fetch = lambda url: FetchResult(
+            status=200,
+            content_type="text/html",
+            text='<html><body><div class="id">99</div></body></html>',
+            response_time_ms=1,
+            content_length=10,
+            final_url=url,
+            headers_dict={},
+            redirect_chain_length=0,
+            fetch_method="static",
+        )
+        fetcher.close = lambda: None
+        return fetcher
+
+    monkeypatch.setattr("website_profiling.crawl.crawler.build_fetcher", fake_build_fetcher)
+
+    c = Crawler(
+        start_url="https://site.com",
+        ignore_robots=True,
+        crawl_auth_username="user",
+        crawl_auth_password="pass",
+        crawl_extra_headers="X-Test: abc\nBadLine",
+        crawl_cookies="sid=1",
+        crawl_robots_txt_override="User-agent: *\nDisallow:",
+        custom_extractors=[{"name": "id", "type": "css", "selector": ".id"}],
+    )
+    assert c.session.headers["Cookie"] == "sid=1"
+    assert c.session.headers["X-Test"] == "abc"
+    assert c.session.auth == ("user", "pass")
+    assert captured["session"] is c.session
+    out = c.worker("https://site.com/page")
+    fields = json.loads(out["custom_fields"])
+    assert fields["id"] == "99"
+
+
+def test_write_and_read_link_edges(monkeypatch) -> None:
+    from website_profiling.db import crawl_store as cs
+
+    conn = CrawlConn(fetchone={"id": 5})
+    monkeypatch.setattr(cs, "get_latest_crawl_run_id", lambda _c: 5)
+    cs.write_link_edges(
+        conn,  # type: ignore[arg-type]
+        [
+            {"from_url": "https://a.com/", "to_url": "https://a.com/b", "anchor_text": "B", "rel": "nofollow"},
+            {"from_url": "", "to_url": "skip"},
+        ],
+        crawl_run_id=5,
+    )
+    assert any("DELETE FROM link_edges" in sql for sql, _ in conn.executed)
+
+    read_conn = CrawlConn(
+        fetchall=[
+            {
+                "from_url": "https://a.com",
+                "to_url": "https://a.com/b",
+                "anchor_text": "B",
+                "rel": "nofollow",
+                "is_nofollow": True,
+                "is_sponsored": False,
+                "is_ugc": False,
+                "link_type": "internal",
+            }
+        ]
+    )
+    rows = cs.read_link_edges(read_conn, run_id=5)  # type: ignore[arg-type]
+    assert rows[0]["anchor_text"] == "B"
+    assert cs.read_link_edges(CrawlConn(boom_execute=True), run_id=5) == []  # type: ignore[arg-type]
+    assert cs.write_link_edges(CrawlConn(), [], crawl_run_id=None) is None  # type: ignore[arg-type]
+    monkeypatch.setattr(cs, "get_latest_crawl_run_id", lambda _c: None)
+    assert cs.write_link_edges(CrawlConn(), [{"from_url": "a", "to_url": "b"}], crawl_run_id=None) is None  # type: ignore[arg-type]
+    assert cs.read_link_edges(CrawlConn(), run_id=None) == []  # type: ignore[arg-type]
+
+
+def test_create_crawl_run_discovery_mode_fallback() -> None:
+    from website_profiling.db import crawl_store as cs
+
+    class BoomDisc(CrawlConn):
+        def execute(self, sql, params=None):
+            self.executed.append((sql, params))
+            if "discovery_mode" in sql:
+                raise RuntimeError("no discovery_mode")
+            return super().execute(sql, params)
+
+    conn = BoomDisc(fetchone={"id": 12})
+    assert cs.create_crawl_run(conn, start_url="https://a.com", discovery_mode="list") == 12  # type: ignore[arg-type]
+
+
+def test_create_crawl_run_raises_when_all_statements_fail() -> None:
+    from website_profiling.db import crawl_store as cs
+
+    conn = CrawlConn(boom_execute=True)
+    with pytest.raises(RuntimeError, match="boom"):
+        cs.create_crawl_run(conn, start_url="https://a.com")  # type: ignore[arg-type]
+
+
+def test_resolve_crawl_user_agent_presets():
+    from website_profiling.crawl.crawler import MOBILE_USER_AGENT, resolve_crawl_user_agent
+
+    assert MOBILE_USER_AGENT in resolve_crawl_user_agent("mobile", None)
+    assert resolve_crawl_user_agent("custom", "MyBot/2") == "MyBot/2"
+
+
+def test_crawler_robots_txt_override_parses(monkeypatch) -> None:
+    from website_profiling.crawl.crawler import Crawler
+
+    monkeypatch.setattr(
+        "website_profiling.crawl.sitemap.discover_sitemap_urls",
+        lambda *_a, **_k: [],
+    )
+    monkeypatch.setattr(
+        "website_profiling.crawl.crawler.load_robots",
+        lambda _u: (_ for _ in ()).throw(AssertionError("load_robots should not run")),
+    )
+    c = Crawler(
+        start_url="https://site.com",
+        ignore_robots=False,
+        crawl_robots_txt_override="User-agent: *\nDisallow: /private",
+    )
+    assert c.rp is not None
+
+
+def test_crawler_loads_robots_when_no_override(monkeypatch) -> None:
+    from website_profiling.crawl.crawler import Crawler
+
+    monkeypatch.setattr(
+        "website_profiling.crawl.sitemap.discover_sitemap_urls",
+        lambda *_a, **_k: [],
+    )
+    monkeypatch.setattr("website_profiling.crawl.crawler.load_robots", lambda _u: object())
+    c = Crawler(start_url="https://site.com", ignore_robots=False)
+    assert c.rp is not None
+
+
+def test_merge_browser_into_page_analysis_sets_axe():
+    from website_profiling.crawl.fetchers.browser_diagnostics import merge_browser_into_page_analysis
+
+    raw = merge_browser_into_page_analysis(
+        "{}",
+        {"axe_violations": [{"id": "label"}]},
+    )
+    assert "axe_violations" in raw
+
+
+def test_run_crawler_writes_link_edges_after_crawl(monkeypatch) -> None:
+    import types
+
+    import pandas as pd
+    import website_profiling.crawl.crawler as mod
+
+    class FakeCrawler:
+        def __init__(self, **_kwargs):
+            self.link_edges_accum = [{"from_url": "https://a.com", "to_url": "https://a.com/b"}]
+
+        def crawl(self, **_kwargs):
+            return pd.DataFrame([{"url": "https://a.com", "status": 200}])
+
+    class _Ctx:
+        def __enter__(self):
+            return object()
+
+        def __exit__(self, _t, _v, _tb):
+            return False
+
+    edge_writes: list[tuple] = []
+
+    fake_db = types.SimpleNamespace(
+        backup_db_if_exists=lambda: None,
+        create_crawl_run=lambda *_a, **_k: 7,
+        db_session=lambda: _Ctx(),
+        read_historical_data=lambda: {},
+        restore_historical_data=lambda *_a, **_k: None,
+        write_crawl=lambda *_a, **_k: None,
+    )
+    fake_storage = types.SimpleNamespace(ensure_crawl_tables_cleared=lambda *_a, **_k: None)
+    monkeypatch.setattr(mod, "Crawler", FakeCrawler)
+    monkeypatch.setitem(__import__("sys").modules, "website_profiling.db", fake_db)
+    monkeypatch.setitem(__import__("sys").modules, "website_profiling.db.storage", fake_storage)
+    monkeypatch.setattr(
+        "website_profiling.db.crawl_store.get_latest_crawl_run_id",
+        lambda _c: 9,
+    )
+    monkeypatch.setattr(
+        "website_profiling.db.crawl_store.write_link_edges",
+        lambda _conn, edges, crawl_run_id=None: edge_writes.append((edges, crawl_run_id)),
+    )
+
+    mod.run_crawler(
+        "https://a.com",
+        output_db=True,
+        preserve_crawl_history=False,
+        show_progress=False,
+    )
+    assert edge_writes
+    assert edge_writes[0][1] == 9
+    assert edge_writes[-1][1] == 7
+
+
+def test_browser_fetcher_passes_auth_headers_to_context(monkeypatch):
+    from website_profiling.crawl.fetchers.browser import BrowserFetcher
+
+    captured: dict = {}
+
+    class CapturingContext:
+        async def new_page(self):
+            from tests.test_browser_fetcher_unit import _FakePage
+
+            return _FakePage()
+
+        async def close(self):
+            return None
+
+    class CapturingBrowser:
+        async def new_context(self, **kwargs):
+            captured.update(kwargs)
+            return CapturingContext()
+
+        async def close(self):
+            return None
+
+    class CapturingChromium:
+        async def launch(self, **_kwargs):
+            return CapturingBrowser()
+
+    class CapturingPlaywright:
+        chromium = CapturingChromium()
+
+        async def stop(self):
+            return None
+
+    class CapturingPlaywrightContext:
+        async def start(self):
+            return CapturingPlaywright()
+
+    fake_api = MagicMock()
+    fake_api.async_playwright = lambda: CapturingPlaywrightContext()
+    monkeypatch.setitem(__import__("sys").modules, "playwright", MagicMock(async_api=fake_api))
+    monkeypatch.setitem(__import__("sys").modules, "playwright.async_api", fake_api)
+
+    fetcher = BrowserFetcher(
+        timeout=5,
+        js_concurrency=1,
+        extra_wait_ms=0,
+        block_resources=False,
+        capture_console=False,
+        extra_http_headers={"Cookie": "sid=1", "X-Auth": "tok"},
+        http_credentials={"username": "u", "password": "p"},
+    )
+    try:
+        fetcher.fetch("https://example.com/")
+        assert captured["extra_http_headers"]["Cookie"] == "sid=1"
+        assert captured["http_credentials"]["username"] == "u"
+    finally:
+        fetcher.close()
+
+
+def test_browser_fetcher_run_axe_attaches_violations(monkeypatch):
+    from website_profiling.crawl.fetchers.browser import BrowserFetcher
+
+    _install_fake_playwright(monkeypatch)
+
+    async def fake_axe(_page):
+        return [{"id": "label", "description": "Missing label", "help": "Add label", "impact": "serious", "nodes": 1}]
+
+    monkeypatch.setattr("website_profiling.crawl.axe_runner.run_axe_on_page", fake_axe)
+
+    fetcher = BrowserFetcher(
+        timeout=5,
+        js_concurrency=1,
+        extra_wait_ms=0,
+        block_resources=False,
+        capture_console=False,
+        run_axe=True,
+    )
+    try:
+        result = fetcher.fetch("https://example.com/")
+        assert result.browser_diagnostics is not None
+        assert result.browser_diagnostics["axe_violations"][0]["id"] == "label"
+    finally:
+        fetcher.close()
diff --git a/tests/test_crawler_deep.py b/tests/test_crawler_deep.py
index fd096d43..fe3e2d99 100644
--- a/tests/test_crawler_deep.py
+++ b/tests/test_crawler_deep.py
@@ -33,7 +33,17 @@ def test_worker_success_path_populates_many_fields(monkeypatch):
         redirect_chain_length=0,
         fetch_method="static",
     )
-    monkeypatch.setattr(mod, "parse_links", lambda _u, _t: ("T", {"https://site.com/a", "https://ext.com/x"}))
+    monkeypatch.setattr(
+        mod,
+        "parse_link_edges",
+        lambda _u, _t: (
+            "T",
+            [
+                {"to_url": "https://site.com/a", "anchor_text": "a", "rel": "", "is_nofollow": False, "is_sponsored": False, "is_ugc": False, "link_type": "internal"},
+                {"to_url": "https://ext.com/x", "anchor_text": "x", "rel": "", "is_nofollow": False, "is_sponsored": False, "is_ugc": False, "link_type": "external"},
+            ],
+        ),
+    )
     monkeypatch.setattr(mod, "parse_seo", lambda *_a, **_k: ("desc", 4, "h1", 1, "https://site.com/canon"))
     monkeypatch.setattr(
         mod,
@@ -412,7 +422,7 @@ def test_run_crawler_non_streaming_db_write(monkeypatch):
 
     class FakeCrawler:
         def __init__(self, **_kwargs):
-            pass
+            self.link_edges_accum = []
 
         def crawl(self, **_kwargs):
             return pd.DataFrame([{"url": "https://a.com", "status": 200, "title": "ok"}])
@@ -428,7 +438,7 @@ def __exit__(self, _t, _v, _tb):
 
     fake_db = types.SimpleNamespace(
         backup_db_if_exists=lambda: "/tmp/backup.sql",
-        create_crawl_run=lambda _c, _u, property_id=None, render_mode=None: 7,
+        create_crawl_run=lambda *_a, **_k: 7,
         db_session=lambda: _Ctx(),
         read_historical_data=lambda: {"report_payload": [{"id": 1}]},
         restore_historical_data=lambda *_a, **_k: None,
@@ -456,7 +466,7 @@ def test_run_crawler_streaming_db_with_history_backup(monkeypatch):
 
     class FakeCrawler:
         def __init__(self, **_kwargs):
-            pass
+            self.link_edges_accum = []
 
         def crawl(self, **_kwargs):
             return pd.DataFrame([{"url": "https://a.com", "status": 200}])
@@ -473,7 +483,7 @@ def __exit__(self, _t, _v, _tb):
 
     fake_db = types.SimpleNamespace(
         backup_db_if_exists=lambda: "/tmp/backup.sql",
-        create_crawl_run=lambda _c, _u, property_id=None, render_mode=None: 11,
+        create_crawl_run=lambda *_a, **_k: 11,
         db_session=lambda: _Ctx(),
         read_historical_data=lambda: {"report_payload": [{"id": 1}]},
         restore_historical_data=lambda *_a, **_k: restored.append(True),
@@ -502,7 +512,7 @@ def test_run_crawler_streaming_db_path(monkeypatch):
 
     class FakeCrawler:
         def __init__(self, **_kwargs):
-            pass
+            self.link_edges_accum = []
 
         def crawl(self, **_kwargs):
             return pd.DataFrame([{"url": "https://a.com", "status": 200}])
@@ -517,7 +527,7 @@ def __exit__(self, _t, _v, _tb):
     monkeypatch.setattr(mod, "Crawler", FakeCrawler)
     fake_db = types.SimpleNamespace(
         backup_db_if_exists=lambda: None,
-        create_crawl_run=lambda _c, _u, property_id=None, render_mode=None: 10,
+        create_crawl_run=lambda *_a, **_k: 10,
         db_session=lambda: _Ctx(),
         read_historical_data=lambda: {},
         restore_historical_data=lambda *_a, **_k: None,
diff --git a/tests/test_crawler_unit.py b/tests/test_crawler_unit.py
index 0dbeb452..b7722384 100644
--- a/tests/test_crawler_unit.py
+++ b/tests/test_crawler_unit.py
@@ -1,3 +1,5 @@
+import json
+
 import pytest
 
 
@@ -524,7 +526,9 @@ def test_worker_respects_exclude_and_max_depth_for_links(monkeypatch) -> None:
     )
     out = c.worker("https://site.com/")
     assert "https://site.com/child" not in c.depths
-    assert out.get("outlink_targets") == "[]"
+    assert "https://site.com/skip/child" not in c.depths
+    targets = json.loads(out.get("outlink_targets") or "[]")
+    assert "https://site.com/child" in targets
 
 
 def test_worker_applies_polite_delay(monkeypatch) -> None:
diff --git a/tests/test_export_sitemap.py b/tests/test_export_sitemap.py
new file mode 100644
index 00000000..56397173
--- /dev/null
+++ b/tests/test_export_sitemap.py
@@ -0,0 +1,34 @@
+"""Tests for sitemap XML export."""
+from __future__ import annotations
+
+from website_profiling.tools.export_sitemap import build_sitemap_xml
+
+
+def test_build_sitemap_xml_includes_urls():
+    payload = {
+        "links": [
+            {"url": "https://example.com/", "status": "200"},
+            {"url": "https://example.com/about", "status": "301"},
+            {"url": "https://example.com/missing", "status": "404"},
+        ]
+    }
+    xml = build_sitemap_xml(payload)
+    assert "<urlset" in xml
+    assert xml.count("<loc>") == 1
+    assert "https://example.com/" in xml
+
+
+def test_build_sitemap_xml_empty_links():
+    assert "<urlset" in build_sitemap_xml({"links": []})
+
+
+def test_build_sitemap_xml_skips_noindex_and_non_dict():
+    payload = {
+        "links": [
+            "bad",
+            {"url": "https://example.com/private", "status": "200", "noindex": True},
+            {"url": "https://example.com/ok", "status": "200"},
+        ]
+    }
+    xml = build_sitemap_xml(payload, max_urls=10)
+    assert xml.count("<loc>") == 1
diff --git a/tests/test_export_workbook.py b/tests/test_export_workbook.py
new file mode 100644
index 00000000..1c8f9c01
--- /dev/null
+++ b/tests/test_export_workbook.py
@@ -0,0 +1,63 @@
+"""Tests for crawl workbook ZIP export."""
+from __future__ import annotations
+
+import zipfile
+import io
+
+from website_profiling.tools.export_crawl_workbook import build_crawl_workbook_zip
+
+
+def test_build_workbook_zip_contains_issues_csv():
+    payload = {
+        "links": [{"url": "https://ex.com/", "status": "200", "title": "Home", "inlinks": 1, "outlinks": 2}],
+        "categories": [{"name": "SEO", "issues": [{"message": "Missing title", "url": "https://ex.com/x", "priority": "High"}]}],
+        "link_edges": [{"from_url": "https://ex.com/", "to_url": "https://ex.com/x", "anchor_text": "x", "rel": "", "is_nofollow": False, "link_type": "internal"}],
+    }
+    raw = build_crawl_workbook_zip(payload)
+    with zipfile.ZipFile(io.BytesIO(raw)) as zf:
+        names = set(zf.namelist())
+    assert "internal_urls.csv" in names
+    assert "issues.csv" in names
+    assert "links.csv" in names
+
+
+def test_build_workbook_zip_custom_fields_columns():
+    payload = {
+        "links": [
+            {
+                "url": "https://ex.com/p",
+                "custom_extract": "SKU-1",
+                "custom_fields": '{"price":"9.99","sku":"SKU-1"}',
+            }
+        ],
+    }
+    raw = build_crawl_workbook_zip(payload)
+    with zipfile.ZipFile(io.BytesIO(raw)) as zf:
+        body = zf.read("custom_fields.csv").decode("utf-8")
+    assert "price" in body
+    assert "9.99" in body
+    assert "SKU-1" in body
+
+
+def test_build_workbook_zip_all_sheets():
+    payload = {
+        "links": [
+            {
+                "url": "https://ex.com/",
+                "status": "200",
+                "title": "Home",
+                "inlinks": 1,
+                "outlinks": 2,
+                "custom_extract": "x",
+                "custom_fields": '{"a":"1"}',
+            }
+        ],
+        "categories": [{"name": "SEO", "issues": [{"message": "x", "url": "https://ex.com/y", "priority": "High"}]}],
+        "link_edges": [{"from_url": "https://ex.com/", "to_url": "https://ex.com/y", "anchor_text": "y", "rel": "", "is_nofollow": False, "link_type": "internal"}],
+        "redirects": [{"url": "https://ex.com/old", "message": "301", "priority": "Low", "recommendation": "fix"}],
+    }
+    raw = build_crawl_workbook_zip(payload)
+    with zipfile.ZipFile(io.BytesIO(raw)) as zf:
+        names = set(zf.namelist())
+    assert "custom_fields.csv" in names
+    assert "redirects.csv" in names
diff --git a/tests/test_extraction.py b/tests/test_extraction.py
new file mode 100644
index 00000000..cef88a70
--- /dev/null
+++ b/tests/test_extraction.py
@@ -0,0 +1,16 @@
+"""Tests for custom extractors."""
+from __future__ import annotations
+
+from website_profiling.crawl.extraction import parse_extractors_config, run_extractors
+
+
+def test_run_css_extractor():
+    html = '<html><div data-id="42">x</div></html>'
+    specs = [{"name": "id", "type": "css", "selector": "[data-id]", "attr": "data-id"}]
+    assert run_extractors(html, specs) == {"id": "42"}
+
+
+def test_parse_extractors_config_json():
+    raw = '[{"name":"a","type":"regex","pattern":"(\\\\d+)"}]'
+    out = parse_extractors_config(raw)
+    assert len(out) == 1 and out[0]["name"] == "a"
diff --git a/tests/test_fix_suggestions.py b/tests/test_fix_suggestions.py
new file mode 100644
index 00000000..c3242675
--- /dev/null
+++ b/tests/test_fix_suggestions.py
@@ -0,0 +1,97 @@
+"""Tests for unified fix suggestion generator."""
+from __future__ import annotations
+
+from unittest.mock import MagicMock, patch
+
+import pytest
+
+from website_profiling.llm.fix_suggestions import (
+    VALID_SOURCES,
+    _build_user_payload,
+    _cache_key,
+    generate_fix_suggestion,
+)
+
+_ENABLED_CFG = {"llm_enabled": True, "llm_enable_issue_fixes": "true", "llm_model": "test-model"}
+
+
+def test_valid_sources() -> None:
+    assert VALID_SOURCES == frozenset(
+        {"issue", "lighthouse", "security", "browser", "seo_content", "technical"}
+    )
+
+
+def test_message_required() -> None:
+    with patch("website_profiling.llm.fix_suggestions.llm_is_enabled", return_value=True):
+        out = generate_fix_suggestion({"source": "lighthouse", "message": "  "}, cfg=_ENABLED_CFG)
+    assert out["ok"] is False
+    assert "message" in out["error"].lower()
+
+
+def test_disabled_llm() -> None:
+    with patch("website_profiling.llm.fix_suggestions.llm_is_enabled", return_value=False):
+        out = generate_fix_suggestion({"message": "Slow LCP"}, cfg=_ENABLED_CFG)
+    assert out["ok"] is False
+    assert "disabled" in out["error"].lower()
+
+
+def test_fix_suggestions_disabled_in_settings() -> None:
+    cfg = {"llm_enabled": True, "llm_enable_issue_fixes": "false"}
+    with patch("website_profiling.llm.fix_suggestions.llm_is_enabled", return_value=True):
+        out = generate_fix_suggestion({"message": "Missing HSTS"}, cfg=cfg)
+    assert out["ok"] is False
+    assert "disabled" in out["error"].lower()
+
+
+def test_cache_key_varies_by_source() -> None:
+    base = {"message": "x", "url": "https://example.com/"}
+    k1 = _cache_key("gpt-4o-mini", "lighthouse", {**base, "source": "lighthouse"})
+    k2 = _cache_key("gpt-4o-mini", "security", {**base, "source": "security"})
+    assert k1 != k2
+
+
+def test_build_user_payload_legacy_issue_fields() -> None:
+    payload = _build_user_payload(
+        {
+            "source": "issue",
+            "message": "Noindex",
+            "url": "https://example.com/a",
+            "priority": "High",
+            "category": "Indexation",
+            "recommendation": "Remove noindex",
+            "type": "noindex",
+        }
+    )
+    assert payload["source"] == "issue"
+    assert payload["context"]["priority"] == "High"
+    assert payload["context"]["category"] == "Indexation"
+
+
+@pytest.mark.parametrize("source", sorted(VALID_SOURCES))
+def test_each_source_accepts_minimal_payload(source: str) -> None:
+    cached_fix = {"fix": "Do the thing.", "effort": "low"}
+    with patch("website_profiling.llm.fix_suggestions.llm_is_enabled", return_value=True):
+        with patch("website_profiling.llm.fix_suggestions._read_cache", return_value=cached_fix):
+            out = generate_fix_suggestion(
+                {"source": source, "message": f"Problem for {source}"},
+                cfg=_ENABLED_CFG,
+            )
+    assert out["ok"] is True
+    assert out["cached"] is True
+    assert out["fix"]["fix"] == "Do the thing."
+
+
+def test_llm_call_writes_cache() -> None:
+    client = MagicMock()
+    client.complete_json.return_value = {"fix": "Add preload.", "effort": "low"}
+    with patch("website_profiling.llm.fix_suggestions.llm_is_enabled", return_value=True):
+        with patch("website_profiling.llm.fix_suggestions._read_cache", return_value=None):
+            with patch("website_profiling.llm.fix_suggestions.get_llm_client", return_value=client):
+                with patch("website_profiling.llm.fix_suggestions._write_cache") as write:
+                    out = generate_fix_suggestion(
+                        {"source": "browser", "message": "TypeError", "context": {"stack": "at foo"}},
+                        cfg=_ENABLED_CFG,
+                    )
+    assert out["ok"] is True
+    assert out["fix"]["fix"] == "Add preload."
+    write.assert_called_once()
diff --git a/tests/test_issue_impact.py b/tests/test_issue_impact.py
new file mode 100644
index 00000000..1ce5975e
--- /dev/null
+++ b/tests/test_issue_impact.py
@@ -0,0 +1,30 @@
+"""Tests for issue impact scoring."""
+from __future__ import annotations
+
+from website_profiling.reporting.issue_impact import (
+    compute_impact_score,
+    enrich_categories_with_traffic_impact,
+    sort_issues_by_impact,
+)
+
+
+def test_compute_impact_score_weights_clicks():
+    assert compute_impact_score("Critical", gsc_clicks=10) > compute_impact_score("Low", gsc_clicks=0)
+
+
+def test_enrich_and_sort_issues():
+    categories = [{
+        "issues": [
+            {"message": "low", "url": "https://ex.com/a", "priority": "Low"},
+            {"message": "high traffic", "url": "https://ex.com/b", "priority": "Medium"},
+        ],
+    }]
+    google = {"gsc": {"pages": [
+        {"page": "https://ex.com/b", "clicks": 50, "impressions": 1000},
+    ]}}
+    enrich_categories_with_traffic_impact(categories, google)
+    issues = categories[0]["issues"]
+    assert issues[1]["gsc_clicks"] == 50
+    assert issues[1]["impact_score"] > issues[0]["impact_score"]
+    sorted_issues = sort_issues_by_impact(issues)
+    assert sorted_issues[0]["message"] == "high traffic"
diff --git a/tests/test_link_edges.py b/tests/test_link_edges.py
new file mode 100644
index 00000000..eca30c84
--- /dev/null
+++ b/tests/test_link_edges.py
@@ -0,0 +1,58 @@
+"""Tests for rich link edge parsing and export fields."""
+from __future__ import annotations
+
+from website_profiling.common import parse_link_edges, parse_links
+
+
+def test_parse_link_edges_anchor_and_rel():
+    html = """
+    <html><head><title>Page</title></head><body>
+      <a href="/about">About us</a>
+      <a href="https://ext.com/x" rel="nofollow sponsored">Partner</a>
+    </body></html>
+    """
+    title, edges = parse_link_edges("https://example.com/", html)
+    assert title == "Page"
+    assert len(edges) == 2
+    internal = next(e for e in edges if e["to_url"].endswith("/about"))
+    assert internal["anchor_text"] == "About us"
+    assert internal["link_type"] == "internal"
+    assert internal["is_nofollow"] is False
+    external = next(e for e in edges if "ext.com" in e["to_url"])
+    assert external["is_nofollow"] is True
+    assert external["is_sponsored"] is True
+    assert external["link_type"] == "external"
+
+
+def test_parse_links_backward_compat():
+    html = '<html><body><a href="/a">A</a><a href="/b">B</a></body></html>'
+    title, links = parse_links("https://example.com", html)
+    assert title == ""
+    assert links == {"https://example.com/a", "https://example.com/b"}
+
+
+def test_workbook_links_csv_columns():
+    from website_profiling.tools.export_crawl_workbook import build_crawl_workbook_zip
+    import zipfile
+    import io
+
+    payload = {
+        "link_edges": [
+            {
+                "from_url": "https://example.com/",
+                "to_url": "https://example.com/about",
+                "anchor_text": "About",
+                "rel": "nofollow",
+                "is_nofollow": True,
+                "is_sponsored": False,
+                "is_ugc": False,
+                "link_type": "internal",
+            }
+        ]
+    }
+    raw = build_crawl_workbook_zip(payload)
+    with zipfile.ZipFile(io.BytesIO(raw)) as zf:
+        header = zf.read("links.csv").decode("utf-8").splitlines()[0]
+    assert "from_url" in header
+    assert "anchor_text" in header
+    assert "is_nofollow" in header
diff --git a/tests/test_mcp_registry.py b/tests/test_mcp_registry.py
index 585c74da..1e853fdb 100644
--- a/tests/test_mcp_registry.py
+++ b/tests/test_mcp_registry.py
@@ -7,7 +7,7 @@
 
 
 def test_tool_definitions_schema() -> None:
-    assert len(TOOL_DEFINITIONS) == 171
+    assert len(TOOL_DEFINITIONS) == 180
     for tool in TOOL_DEFINITIONS:
         assert tool.get("name")
         assert tool.get("description")
diff --git a/tests/test_mcp_server_helpers.py b/tests/test_mcp_server_helpers.py
index 01a73b2a..1a5dfaed 100644
--- a/tests/test_mcp_server_helpers.py
+++ b/tests/test_mcp_server_helpers.py
@@ -61,7 +61,7 @@ def test_read_glossary_excerpt_missing(monkeypatch) -> None:
 
 def test_tools_catalog_json_includes_security_tools() -> None:
     catalog = json.loads(mcp_server._tools_catalog_json())
-    assert catalog["tool_count"] >= 171
+    assert catalog["tool_count"] >= 176
     assert "get_security_findings" in catalog["domains"]["security"]
 
 
@@ -168,7 +168,7 @@ async def __aexit__(self, *_args):
     assert captured["name"] == "site-audit"
     assert captured["ran"] is True
     tools = asyncio.run(captured["list_tools"]())  # type: ignore[arg-type]
-    assert len(tools) >= 171
+    assert len(tools) >= 176
     resources = asyncio.run(captured["list_resources"]())  # type: ignore[arg-type]
     assert any(r["uri"] == "audit://property/7" for r in resources)
 
@@ -186,5 +186,11 @@ def test_mcp_package_main(monkeypatch) -> None:
 
 
 def test_mcp_server_main_guard() -> None:
+    # run_module executes __main__ in a fresh import; drop any prior import from this file.
+    sys.modules.pop("website_profiling.mcp.server", None)
     with pytest.raises(SystemExit, match="MCP SDK"):
-        runpy.run_module("website_profiling.mcp.server", run_name="__main__")
+        runpy.run_module(
+            "website_profiling.mcp.server",
+            run_name="__main__",
+            alter_sys=False,
+        )
diff --git a/tests/test_optional_audits.py b/tests/test_optional_audits.py
new file mode 100644
index 00000000..99dfd643
--- /dev/null
+++ b/tests/test_optional_audits.py
@@ -0,0 +1,49 @@
+"""Tests for optional audit hooks."""
+from __future__ import annotations
+
+import json
+
+import pandas as pd
+
+from website_profiling.reporting.optional_audits import (
+    amp_audit_issues,
+    apply_optional_audits,
+    pagination_issues,
+)
+
+
+def test_pagination_issues_from_page_analysis():
+    df = pd.DataFrame([
+        {
+            "url": "https://example.com/page/2",
+            "status": "200",
+            "page_analysis": json.dumps({"pagination": {"rel_prev": "https://example.com/page/1", "rel_next": None}}),
+        },
+    ])
+    issues = pagination_issues(df)
+    assert any("rel=prev" in i["message"] for i in issues)
+
+
+def test_amp_audit_missing_canonical():
+    df = pd.DataFrame([
+        {
+            "url": "https://example.com/amp/article",
+            "status": "200",
+            "canonical_url": "",
+            "content_type": "text/html",
+            "page_analysis": "{}",
+        },
+    ])
+    issues = amp_audit_issues(df)
+    assert len(issues) == 1
+
+
+def test_apply_optional_audits_spell_skipped_without_package():
+    categories = [
+        {"id": "technical_seo", "name": "Technical", "issues": [], "recommendations": []},
+        {"id": "intelligence", "name": "Content", "issues": [], "recommendations": []},
+        {"id": "html_accessibility", "name": "A11y", "issues": [], "recommendations": []},
+    ]
+    df = pd.DataFrame([{"url": "https://example.com", "status": "200", "page_analysis": "{}"}])
+    meta = apply_optional_audits(categories, df, {"enable_spell_check": "true"})
+    assert isinstance(meta, dict)
diff --git a/tests/test_pipeline_report_pool_unit.py b/tests/test_pipeline_report_pool_unit.py
index cfeecc43..d92af4de 100644
--- a/tests/test_pipeline_report_pool_unit.py
+++ b/tests/test_pipeline_report_pool_unit.py
@@ -89,6 +89,35 @@ def fake_run_plot(**kwargs):
     assert captured.get("js_block_resources") is False
 
 
+def test_run_plot_emits_done_on_failure(monkeypatch):
+    from website_profiling.commands import pipeline_cmd
+
+    done_calls: list[tuple[str, dict]] = []
+
+    monkeypatch.setattr(pipeline_cmd, "emit_phase_start", lambda *_a, **_k: None)
+    monkeypatch.setattr(
+        pipeline_cmd,
+        "emit_phase_done",
+        lambda phase, message=None: done_calls.append((phase, {"message": message})),
+    )
+
+    def boom(**_kwargs):
+        raise RuntimeError("plot failed")
+
+    monkeypatch.setitem(
+        __import__("sys").modules,
+        "website_profiling.tools.plot",
+        types.SimpleNamespace(run_plot=boom),
+    )
+
+    import pytest
+
+    with pytest.raises(RuntimeError, match="plot failed"):
+        pipeline_cmd._run_plot({}, True)
+
+    assert done_calls == [("plot", {"message": "Charts step failed"})]
+
+
 def test_build_report_metadata_includes_fetch_method_counts() -> None:
     import pandas as pd
     from website_profiling.reporting.builder import _build_report_metadata
diff --git a/tests/test_progress.py b/tests/test_progress.py
new file mode 100644
index 00000000..22849bc9
--- /dev/null
+++ b/tests/test_progress.py
@@ -0,0 +1,51 @@
+"""Tests for structured pipeline progress events."""
+from __future__ import annotations
+
+import json
+from unittest.mock import patch
+
+from website_profiling.progress import CrawlProgressTracker, PREFIX, emit_progress, emit_phase_done
+
+
+def test_emit_progress_prints_json_line(capsys):
+    emit_progress("crawl", "fetch", current=3, total=10, url="https://ex.com/a")
+    out = capsys.readouterr().out.strip()
+    assert out.startswith(PREFIX)
+    payload = json.loads(out[len(PREFIX) :])
+    assert payload["phase"] == "crawl"
+    assert payload["step"] == "fetch"
+    assert payload["current"] == 3
+    assert payload["total"] == 10
+    assert payload["url"] == "https://ex.com/a"
+    assert "ts" in payload
+
+
+def test_crawl_tracker_throttles(capsys):
+    tracker = CrawlProgressTracker(total=100, start_time=1000.0)
+    with patch("website_profiling.progress.time") as mock_time:
+        # Each emit calls time() twice (tracker + payload ts).
+        mock_time.time.side_effect = [
+            1000.0, 1000.0,
+            1000.5,
+            1000.6,
+            1003.0, 1003.0,
+            1008.0, 1008.0,
+        ]
+        tracker.maybe_emit(1, "https://a.com")
+        tracker.maybe_emit(2, "https://b.com")
+        tracker.maybe_emit(3, "https://c.com")
+        tracker.maybe_emit(6, "https://f.com")
+    lines = [ln for ln in capsys.readouterr().out.splitlines() if ln.startswith(PREFIX)]
+    assert len(lines) == 2
+    last = json.loads(lines[-1][len(PREFIX) :])
+    assert last["current"] == 6
+
+
+def test_emit_phase_done():
+    with patch("website_profiling.progress.print") as mock_print:
+        emit_phase_done("report")
+        args = mock_print.call_args[0][0]
+        assert args.startswith(PREFIX)
+        payload = json.loads(args[len(PREFIX) :])
+        assert payload["phase"] == "report"
+        assert payload["step"] == "done"
diff --git a/tests/test_property_profile.py b/tests/test_property_profile.py
new file mode 100644
index 00000000..684d9e41
--- /dev/null
+++ b/tests/test_property_profile.py
@@ -0,0 +1,559 @@
+"""Tests for property profile: site files, subdomains, contact intelligence."""
+from __future__ import annotations
+
+import json
+from unittest.mock import MagicMock, patch
+
+import pandas as pd
+
+from website_profiling.analysis.page import (
+    _append_contact,
+    _collect_json_ld_contacts,
+    _collect_json_ld_types,
+    _format_postal_address,
+    _normalize_type_name,
+    analyze_html,
+)
+from website_profiling.reporting.builder import _fetch_site_level
+from website_profiling.reporting.categories import category_technical_seo, merge_subdomain_issues
+from website_profiling.reporting.contact_intelligence import (
+    _apex_from_start_url,
+    _has_org_schema,
+    _merge_entry,
+    _parse_page_analysis,
+    _signals_from_page,
+    build_contact_intelligence,
+)
+from website_profiling.reporting.site_files import (
+    fetch_ads_txt,
+    fetch_rdap_org_name,
+    fetch_security_txt,
+    merge_site_file_fields,
+    parse_ads_txt,
+    parse_security_txt,
+)
+from website_profiling.reporting.subdomains import (
+    _crawl_hosts,
+    _fetch_crtsh_hosts,
+    _gsc_hosts,
+    _host_in_scope,
+    build_subdomain_inventory,
+)
+
+
+def test_parse_ads_txt_valid_lines():
+    text = "example.com, pub-123, DIRECT\n# comment\n\nbad-line"
+    out = parse_ads_txt(text)
+    assert out["ads_txt_present"] is True
+    assert out["ads_txt_line_count"] == 1
+    assert out["ads_txt_valid"] is False
+    assert "invalid_line:4" in out["ads_txt_issues"]
+
+
+def test_parse_ads_txt_empty():
+    out = parse_ads_txt("")
+    assert out["ads_txt_present"] is False
+    assert out["ads_txt_valid"] is False
+
+
+def test_parse_security_txt_contact_and_expires():
+    text = "Contact: mailto:sec@example.com\nExpires: 2030-01-01T00:00:00Z\n"
+    out = parse_security_txt(text)
+    assert out["security_txt_present"] is True
+    assert out["security_txt_valid"] is True
+    assert out["security_txt_contact"] == ["mailto:sec@example.com"]
+    assert out["security_txt_expires"] == "2030-01-01T00:00:00Z"
+
+
+def test_json_ld_contact_helper_edge_cases():
+    assert _normalize_type_name(["Organization", "Person"]) == "Organization"
+    assert _normalize_type_name(42) == ""
+
+    types: set[str] = set()
+    _collect_json_ld_types(
+        {"@type": "WebPage", "mainEntity": {"@type": "Article", "about": [{"@type": "Thing"}]}},
+        types,
+    )
+    assert "WebPage" in types
+    assert "Article" in types
+    assert "Thing" in types
+
+    assert _format_postal_address(" 123 Main St ") == "123 Main St"
+    assert _format_postal_address(99) == ""
+    assert (
+        _format_postal_address(
+            {
+                "streetAddress": "1 Main",
+                "addressLocality": "Town",
+                "addressRegion": "CA",
+                "postalCode": "90210",
+                "addressCountry": "US",
+            }
+        )
+        == "1 Main, Town, CA, 90210, US"
+    )
+
+    signals: dict[str, list[str]] = {}
+    _append_contact(signals, "emails", 123)
+    _append_contact(signals, "emails", "   ")
+    _append_contact(signals, "emails", "a@x.com")
+    _append_contact(signals, "emails", "a@x.com")
+    assert signals["emails"] == ["a@x.com"]
+
+    for i in range(12):
+        _append_contact(signals, "phones", f"+1{i:03d}")
+    assert len(signals["phones"]) == 10
+
+    contact_payload = {
+        "@type": ["LocalBusiness", "Organization"],
+        "name": "Shop",
+        "email": "shop@example.com",
+        "telephone": "+1-555-0100",
+        "address": "123 Simple St",
+        "@graph": [
+            {
+                "@type": "NewsMediaOrganization",
+                "name": "Press",
+                "email": "press@example.com",
+                "telephone": "+1-555-0200",
+            },
+            {
+                "subsidiary": {
+                    "@type": "Organization",
+                    "name": "Nested Org",
+                    "email": "nested@example.com",
+                    "address": {"streetAddress": "9 Side", "addressLocality": "Alley"},
+                }
+            },
+        ],
+    }
+    org_signals: dict[str, list[str]] = {}
+    _collect_json_ld_contacts(contact_payload, org_signals)
+    assert "shop@example.com" in org_signals["emails"]
+    assert "press@example.com" in org_signals["emails"]
+    assert "nested@example.com" in org_signals["emails"]
+    assert "123 Simple St" in org_signals["addresses"]
+    assert "9 Side, Alley" in org_signals["addresses"]
+
+    list_signals: dict[str, list[str]] = {}
+    _collect_json_ld_contacts(
+        [{"@type": "Organization", "name": "List Org", "email": "list@example.com"}],
+        list_signals,
+    )
+    assert "list@example.com" in list_signals["emails"]
+
+
+def test_analyze_html_contact_and_json_ld_types():
+    html = """
+    <html><head>
+    <script type="application/ld+json">
+    {"@type":"Organization","name":"Acme Inc","email":"hello@acme.com","telephone":"+1-555-0100"}
+    </script>
+    </head><body>
+    <a href="mailto:sales@acme.com">Email</a>
+    <a href="tel:+15550199">Call</a>
+    </body></html>
+    """
+    out = analyze_html(html, "https://acme.com/", "https://acme.com/")
+    assert "Organization" in out.get("json_ld_types", [])
+    signals = out.get("contact_signals") or {}
+    assert "hello@acme.com" in signals.get("emails", [])
+    assert "sales@acme.com" in signals.get("emails", [])
+    assert any("+1555" in p for p in signals.get("phones", []))
+
+
+def test_build_contact_intelligence_dedup_and_security_txt():
+    pa = json.dumps(
+        {
+            "json_ld_types": ["Organization"],
+            "contact_signals": {
+                "emails": ["Hello@Example.com", "other@example.com"],
+                "phones": [],
+                "addresses": [],
+                "organization_names": ["Example Co"],
+            },
+        }
+    )
+    df = pd.DataFrame([{"url": "https://example.com/contact", "status": "200", "page_analysis": pa}])
+    site_level = {"security_txt_contact": ["mailto:sec@example.com"]}
+    with patch("website_profiling.reporting.contact_intelligence.fetch_rdap_org_name", return_value=None):
+        out = build_contact_intelligence(df, site_level, "https://example.com/", {"enable_rdap_org_lookup": "false"})
+    emails = {e["value"].lower() for e in out["emails"]}
+    assert "hello@example.com" in emails
+    assert "other@example.com" in emails
+    assert "sec@example.com" in emails
+    assert out["primary_contact_page"] == "https://example.com/contact"
+
+
+def test_build_subdomain_inventory_crawl_and_gsc():
+    df = pd.DataFrame(
+        [
+            {"url": "https://www.example.com/", "status": "200"},
+            {"url": "https://www.example.com/about", "status": "200"},
+        ]
+    )
+    indexation = {
+        "lists": {"gsc_not_crawled": ["https://blog.example.com/post"]},
+        "url_join": {},
+    }
+    with patch("website_profiling.reporting.subdomains._fetch_crtsh_hosts", return_value=(set(), None)):
+        out = build_subdomain_inventory(df, indexation, "https://www.example.com/", {"subdomain_ct_lookup": "false"})
+    hosts = {h["host"]: h for h in out["hosts"]}
+    assert "www.example.com" in hosts
+    assert hosts["www.example.com"]["in_crawl"] is True
+    assert "blog.example.com" in out["gsc_hosts_not_crawled"]
+
+
+def test_merge_subdomain_issues_adds_medium_issue():
+    categories = [{"id": "technical_seo", "issues": [], "recommendations": []}]
+    merge_subdomain_issues(categories, {"gsc_hosts_not_crawled": ["blog.example.com"]})
+    assert len(categories[0]["issues"]) == 1
+    assert categories[0]["issues"][0]["priority"] == "Medium"
+
+
+def test_category_technical_seo_missing_ads_and_security_txt():
+    df = pd.DataFrame([{"url": "https://example.com", "status": "200"}])
+    site_level = {
+        "robots_present": True,
+        "sitemap_present": True,
+        "sitemap_valid": True,
+        "ads_txt_present": False,
+        "security_txt_present": False,
+    }
+    cat = category_technical_seo(df, site_level)
+    msgs = " ".join(i["message"] for i in cat["issues"])
+    assert "ads.txt" in msgs
+    assert "security.txt" in msgs
+
+
+def test_parse_ads_txt_no_valid_sellers():
+    out = parse_ads_txt("# only comments\ninvalid-line\n")
+    assert out["ads_txt_present"] is True
+    assert out["ads_txt_valid"] is False
+    assert "no_sellers" in out["ads_txt_issues"]
+
+
+def test_parse_security_txt_skips_blank_and_malformed_lines():
+    text = "\n# comment\nno-colon-line\nContact:\nContact: mailto:ok@example.com\n"
+    out = parse_security_txt(text)
+    assert out["security_txt_contact"] == ["mailto:ok@example.com"]
+
+
+def test_fetch_security_txt_falls_back_and_handles_errors():
+    class FakeResp:
+        def __init__(self, status_code, text):
+            self.status_code = status_code
+            self.text = text
+
+    session = MagicMock()
+    session.get.side_effect = [
+        FakeResp(404, ""),
+        RuntimeError("timeout"),
+    ]
+    assert fetch_security_txt(session, "https://example.com")["security_txt_present"] is False
+
+
+def test_fetch_ads_and_security_txt_helpers():
+    class FakeResp:
+        def __init__(self, status_code, text):
+            self.status_code = status_code
+            self.text = text
+
+    session = MagicMock()
+    session.get.side_effect = [
+        FakeResp(404, ""),
+        FakeResp(200, "Contact: mailto:sec@example.com\n"),
+    ]
+    assert fetch_ads_txt(session, "https://example.com")["ads_txt_present"] is False
+    sec = fetch_security_txt(session, "https://example.com")
+    assert sec["security_txt_present"] is True
+
+    session.get.side_effect = RuntimeError("network")
+    assert fetch_ads_txt(session, "https://example.com")["ads_txt_present"] is False
+
+
+def test_fetch_rdap_org_name_paths(monkeypatch):
+    class FakeResp:
+        def __init__(self, status_code, payload):
+            self.status_code = status_code
+            self._payload = payload
+
+        def json(self):
+            return self._payload
+
+    monkeypatch.setattr(
+        "website_profiling.reporting.site_files.requests.get",
+        lambda *a, **k: FakeResp(
+            200,
+            {
+                "entities": [
+                    "bad",
+                    {
+                        "roles": ["administrative"],
+                        "vcardArray": ["vcard", ["org", {}, "text", "Ignored"]],
+                    },
+                    {"roles": ["registrant"], "vcardArray": ["vcard"]},
+                    {"roles": ["registrant"], "vcardArray": ["vcard", ["fn", {}, "text", ""]]},
+                    {"roles": ["registrant"], "vcardArray": ["vcard", ["bad"]]},
+                    {
+                        "roles": ["registrant"],
+                        "vcardArray": ["vcard", ["org", {}, "text", "Org Name Inc"]],
+                    },
+                ]
+            },
+        ),
+    )
+    assert fetch_rdap_org_name("www.example.com") == "Org Name Inc"
+
+    monkeypatch.setattr(
+        "website_profiling.reporting.site_files.requests.get",
+        lambda *a, **k: FakeResp(
+            200,
+            {
+                "entities": [
+                    {
+                        "roles": ["registrant"],
+                        "vcardArray": ["vcard", ["fn", {}, "text", "Acme Holdings"]],
+                    }
+                ]
+            },
+        ),
+    )
+    assert fetch_rdap_org_name("example.com") == "Acme Holdings"
+    assert fetch_rdap_org_name("") is None
+    assert fetch_rdap_org_name("localhost") is None
+
+    monkeypatch.setattr(
+        "website_profiling.reporting.site_files.requests.get",
+        lambda *a, **k: FakeResp(404, {}),
+    )
+    assert fetch_rdap_org_name("example.com") is None
+
+    monkeypatch.setattr(
+        "website_profiling.reporting.site_files.requests.get",
+        lambda *a, **k: FakeResp(200, {"entities": "bad"}),
+    )
+    assert fetch_rdap_org_name("example.com") is None
+
+    monkeypatch.setattr(
+        "website_profiling.reporting.site_files.requests.get",
+        lambda *a, **k: (_ for _ in ()).throw(RuntimeError("rdap down")),
+    )
+    assert fetch_rdap_org_name("example.com") is None
+
+    monkeypatch.setattr(
+        "website_profiling.reporting.site_files.requests.get",
+        lambda *a, **k: FakeResp(200, {"entities": [{"roles": ["admin"], "vcardArray": ["vcard", ["fn", {}, "text", "X"]]}]}),
+    )
+    assert fetch_rdap_org_name("example.com") is None
+
+
+def test_merge_site_file_fields():
+    out: dict = {}
+    merge_site_file_fields(out, {"ads_txt_present": True, "security_txt_present": False})
+    assert out["ads_txt_present"] is True
+
+
+def test_contact_intelligence_helpers_and_branches():
+    assert _parse_page_analysis({"emails": ["a@x.com"]})["emails"] == ["a@x.com"]
+    assert _parse_page_analysis(float("nan")) == {}
+    assert _parse_page_analysis("{}") == {}
+    assert _parse_page_analysis("{bad") == {}
+
+    bucket: dict = {}
+    _merge_entry(bucket, "   ", source="x")
+    _merge_entry(bucket, "A@X.com", source="json_ld", url="https://example.com")
+    _merge_entry(bucket, "a@x.com", source="tel_link", url="https://example.com/contact")
+    assert bucket["a@x.com"]["sources"] == ["json_ld", "tel_link"]
+
+    assert _signals_from_page({"contact_signals": "bad"}) == {}
+    assert _signals_from_page({"contact_signals": {"emails": [" a@x.com ", ""]}})["emails"] == ["a@x.com"]
+    assert _has_org_schema({"json_ld_types": "Organization"}) is True
+    assert _has_org_schema({"json_ld_types": 42}) is False
+    assert _apex_from_start_url("https://www.example.com/") == "example.com"
+
+
+def test_build_contact_intelligence_rich_paths():
+    pa_home = json.dumps(
+        {
+            "json_ld_types": ["Organization"],
+            "contact_signals": {
+                "emails": ["home@example.com"],
+                "phones": ["+1-555-0001"],
+                "addresses": ["1 Main"],
+                "organization_names": ["Example Co"],
+            },
+        }
+    )
+    pa_about = json.dumps(
+        {
+            "contact_signals": {
+                "emails": ["about@example.com", "extra1@example.com", "extra2@example.com", "extra3@example.com"],
+                "organization_names": ["Other Co"],
+            }
+        }
+    )
+    df = pd.DataFrame(
+        [
+            {"url": "https://www.example.com/", "status": "200", "page_analysis": pa_home},
+            {"url": "https://www.example.com/about", "status": "200", "page_analysis": pa_about},
+            {"url": "", "status": "200", "page_analysis": "{}"},
+        ]
+    )
+    site_level = {
+        "security_txt_contact": [
+            "mailto:sec@example.com",
+            "tel:+15550100",
+            "security@example.com",
+            42,
+        ]
+    }
+    with patch("website_profiling.reporting.contact_intelligence.fetch_rdap_org_name", return_value="RDAP Org"):
+        out = build_contact_intelligence(df, site_level, "https://www.example.com/", {"enable_rdap_org_lookup": "true"})
+    assert out["primary_contact_page"] == "https://www.example.com/about"
+    assert any(n["value"] == "RDAP Org" for n in out["organization_names"])
+    assert any("distinct email" in n for n in out["consistency_notes"])
+    assert any("organization names" in n for n in out["consistency_notes"])
+    assert not any("homepage" in n.lower() for n in out["consistency_notes"])
+
+
+def test_build_contact_intelligence_without_home_org_schema():
+    pa = json.dumps({"contact_signals": {"emails": ["only@example.com"]}})
+    df = pd.DataFrame([{"url": "https://example.com/page", "status": "200", "page_analysis": pa}])
+    with patch("website_profiling.reporting.contact_intelligence.fetch_rdap_org_name", return_value=None):
+        out = build_contact_intelligence(df, {}, "https://example.com/", {"enable_rdap_org_lookup": "false"})
+    assert out["primary_contact_page"] == "https://example.com/page"
+    assert any("homepage" in n.lower() for n in out["consistency_notes"])
+
+
+def test_subdomain_inventory_helpers_and_ct(monkeypatch):
+    assert _host_in_scope("", "example.com") is False
+    assert _fetch_crtsh_hosts("")[0] == set()
+    assert _crawl_hosts(pd.DataFrame()) == {}
+    assert _gsc_hosts(None) == ({}, [])
+    counts, not_crawled = _gsc_hosts(
+        {
+            "lists": {"gsc_not_crawled": ["https://blog.example.com/post"]},
+            "url_join": {"gsc_only": [{"url": "https://shop.example.com/p"}]},
+        }
+    )
+    assert counts["blog.example.com"] == 1
+    assert counts["shop.example.com"] == 1
+    assert "blog.example.com" in not_crawled
+
+    class FakeResp:
+        status_code = 200
+
+        @staticmethod
+        def json():
+            return [{"name_value": "api.example.com\n*.example.com"}]
+
+    monkeypatch.setattr("website_profiling.reporting.subdomains.requests.get", lambda *a, **k: FakeResp())
+    hosts, err = _fetch_crtsh_hosts("example.com")
+    assert "api.example.com" in hosts
+    assert err is None
+
+    disabled = build_subdomain_inventory(pd.DataFrame(), None, "https://www.example.com/", {"enable_subdomain_discovery": "false"})
+    assert disabled["disabled"] is True
+
+    df = pd.DataFrame([{"url": "https://www.example.com/", "status": "200"}])
+    indexation = {"lists": {"gsc_not_crawled": ["https://blog.example.com/post"]}, "url_join": {}}
+    with patch("website_profiling.reporting.subdomains._fetch_crtsh_hosts", return_value=({"cdn.example.com"}, None)):
+        out = build_subdomain_inventory(df, indexation, "https://www.example.com/", {"subdomain_ct_lookup": "true"})
+    sources = {h["host"]: h["sources"] for h in out["hosts"]}
+    assert "crtsh" in sources["cdn.example.com"]
+
+    with patch("website_profiling.reporting.subdomains._fetch_crtsh_hosts", return_value=(set(), "crtsh: timeout")):
+        err_out = build_subdomain_inventory(df, indexation, "https://www.example.com/", {})
+    assert err_out["crtsh_error"] == "crtsh: timeout"
+
+    class BadStatus:
+        status_code = 503
+
+    monkeypatch.setattr("website_profiling.reporting.subdomains.requests.get", lambda *a, **k: BadStatus())
+    hosts, err = _fetch_crtsh_hosts("example.com")
+    assert hosts == set()
+    assert "HTTP 503" in (err or "")
+
+    class BadJson:
+        status_code = 200
+
+        @staticmethod
+        def json():
+            return {"not": "a list"}
+
+    monkeypatch.setattr("website_profiling.reporting.subdomains.requests.get", lambda *a, **k: BadJson())
+    hosts, err = _fetch_crtsh_hosts("example.com")
+    assert "unexpected response" in (err or "")
+
+    class RowJson:
+        status_code = 200
+
+        @staticmethod
+        def json():
+            return ["bad", {"name_value": "ok.example.com"}]
+
+    monkeypatch.setattr("website_profiling.reporting.subdomains.requests.get", lambda *a, **k: RowJson())
+    hosts, _ = _fetch_crtsh_hosts("example.com")
+    assert "ok.example.com" in hosts
+
+    monkeypatch.setattr(
+        "website_profiling.reporting.subdomains.requests.get",
+        lambda *a, **k: (_ for _ in ()).throw(RuntimeError("crtsh down")),
+    )
+    hosts, err = _fetch_crtsh_hosts("example.com")
+    assert "crtsh down" in (err or "")
+
+    counts, _ = _gsc_hosts(
+        {
+            "lists": {},
+            "url_join": {
+                "gsc_only": [
+                    {"page": "https://shop.example.com/item"},
+                    "https://plain.example.com/x",
+                    "",
+                ]
+            },
+        }
+    )
+    assert counts["shop.example.com"] == 1
+    assert counts["plain.example.com"] == 1
+
+
+def test_merge_subdomain_issues_edge_cases():
+    categories = [{"id": "technical_seo", "issues": [], "recommendations": []}]
+    merge_subdomain_issues(categories, {"disabled": True})
+    merge_subdomain_issues(categories, {"gsc_hosts_not_crawled": []})
+    assert categories[0]["issues"] == []
+
+    merge_subdomain_issues(
+        categories,
+        {"gsc_hosts_not_crawled": [f"host{i}.example.com" for i in range(7)]},
+    )
+    assert len(categories[0]["issues"]) == 1
+    assert "(+2 more)" in categories[0]["issues"][0]["message"]
+
+
+def test_fetch_site_level_includes_ads_and_security(monkeypatch):
+    class FakeResp:
+        def __init__(self, status_code, text):
+            self.status_code = status_code
+            self.text = text
+
+    def fake_get(url, timeout=8):
+        if url.endswith("/ads.txt"):
+            return FakeResp(200, "example.com, pub-1, DIRECT")
+        if url.endswith("/.well-known/security.txt"):
+            return FakeResp(200, "Contact: mailto:sec@example.com")
+        if url.endswith("/robots.txt"):
+            return FakeResp(200, "User-agent: *\n")
+        if url.endswith("/sitemap.xml"):
+            return FakeResp(200, '<?xml version="1.0"?><urlset></urlset>')
+        return FakeResp(404, "")
+
+    session = MagicMock()
+    session.get.side_effect = fake_get
+    monkeypatch.setattr("website_profiling.reporting.builder.requests.Session", lambda: session)
+    out = _fetch_site_level("https://example.com/", timeout=1)
+    assert out["ads_txt_present"] is True
+    assert out["security_txt_present"] is True
diff --git a/tests/test_report_categories_golden.py b/tests/test_report_categories_golden.py
index 8845e7f8..c5347f0a 100644
--- a/tests/test_report_categories_golden.py
+++ b/tests/test_report_categories_golden.py
@@ -66,3 +66,28 @@ def test_build_categories_golden_fingerprints() -> None:
 
     ids = {c["id"] for c in categories}
     assert ids >= {"technical_seo", "core_web_vitals", "link_health", "security", "performance"}
+
+
+def test_build_categories_missing_site_files_low_issues() -> None:
+    rows = json.loads((FIXTURES / "minimal_crawl.json").read_text(encoding="utf-8"))
+    df = pd.DataFrame(rows)
+    site_level = {
+        "robots_present": True,
+        "sitemap_present": True,
+        "sitemap_valid": True,
+        "ads_txt_present": False,
+        "security_txt_present": False,
+    }
+    categories = build_categories(
+        df,
+        [],
+        {"issues": {}},
+        site_level,
+        "https://example.com/",
+    )
+    tech = next(c for c in categories if c.get("id") == "technical_seo")
+    msgs = " ".join(str(i.get("message") or "") for i in tech.get("issues") or [])
+    assert "ads.txt" in msgs.lower()
+    assert "security.txt" in msgs.lower()
+    priorities = {str(i.get("priority")) for i in tech.get("issues") or [] if "ads.txt" in str(i.get("message") or "").lower()}
+    assert "Low" in priorities
diff --git a/tests/test_reporting_gaps.py b/tests/test_reporting_gaps.py
new file mode 100644
index 00000000..bee58784
--- /dev/null
+++ b/tests/test_reporting_gaps.py
@@ -0,0 +1,312 @@
+"""Reporting module coverage for issue impact, link edges, and optional audits."""
+from __future__ import annotations
+
+import json
+from unittest.mock import MagicMock
+
+import pandas as pd
+
+from website_profiling.reporting.issue_impact import (
+    compute_impact_score,
+    enrich_categories_with_traffic_impact,
+    sort_issues_by_impact,
+)
+from website_profiling.reporting.link_edges_report import build_inlink_anchor_matrix, summarize_link_rel
+from website_profiling.reporting.optional_audits import (
+    amp_audit_issues,
+    apply_optional_audits,
+    axe_issues_from_df,
+    html_validation_issues,
+    pagination_issues,
+    spell_check_issues,
+    wayback_issues,
+)
+
+LONG_HTML = "<html><head><title>A</title><title>B</title></head><body>" + ("x" * 120) + "</body></html>"
+
+
+def test_issue_impact_enriches_and_sorts():
+    categories = [
+        {
+            "id": "seo",
+            "issues": [
+                {"url": "https://Example.com/Page", "priority": "High"},
+                {"url": "https://other.com/", "priority": "Low"},
+            ],
+        }
+    ]
+    google = {
+        "gsc": {"top_pages": [{"page": "https://example.com/page", "clicks": 3, "impressions": 100}]},
+        "ga4": {"pages": [{"path": "/page", "sessions": 2}]},
+    }
+    enrich_categories_with_traffic_impact(categories, google)
+    assert categories[0]["issues"][0]["impact_score"] > categories[0]["issues"][1]["impact_score"]
+    assert categories[0]["issues"][0]["gsc_clicks"] == 3
+    sorted_issues = sort_issues_by_impact(categories[0]["issues"])
+    assert sorted_issues[0]["url"].startswith("https://Example.com")
+
+
+def test_compute_impact_score_defaults():
+    assert compute_impact_score("Unknown") >= 1
+
+
+def test_issue_impact_handles_invalid_rows():
+    assert enrich_categories_with_traffic_impact([], []) == []
+    enrich_categories_with_traffic_impact(
+        [
+            "bad",
+            {
+                "issues": [
+                    "bad-issue",
+                    {"url": "", "priority": "Low"},
+                    {"url": "https://example.com/x", "priority": "Medium"},
+                ]
+            },
+        ],
+        {
+            "gsc": {"top_pages": ["bad", {"page": "", "clicks": 1}, {"page": "https://example.com/x", "clicks": 2, "impressions": 5}]},
+            "ga4": {"pages": ["bad", {"path": "", "sessions": 1}, {"path": "/x", "sessions": 4}]},
+        },
+    )
+    assert sort_issues_by_impact(
+        [{"priority": "High", "impact_score": 5}, {"priority": "Low", "impact_score": 1}]
+    )[0]["impact_score"] == 5
+
+
+def test_link_edges_matrix_skips_incomplete_edges():
+    rows = build_inlink_anchor_matrix(
+        [{"from_url": "", "to_url": "https://a.com/x", "anchor_text": "X", "link_type": "internal"}]
+    )
+    assert rows == []
+
+
+def test_link_edges_report_summaries():
+    edges = [
+        {"from_url": "https://a.com/", "to_url": "https://a.com/x", "anchor_text": "X", "link_type": "internal", "is_nofollow": True, "is_sponsored": False, "is_ugc": False},
+        {"from_url": "https://a.com/", "to_url": "https://ext.com/", "anchor_text": "Ext", "link_type": "external"},
+    ]
+    summary = summarize_link_rel(edges)
+    assert summary["total_edges"] == 2
+    assert summary["nofollow_internal"] == 1
+    matrix = build_inlink_anchor_matrix(edges)
+    assert matrix[0]["target_url"] == "https://a.com/x"
+    assert matrix[0]["inlink_count"] == 1
+
+
+def test_pagination_amp_mismatch():
+    df = pd.DataFrame([
+        {
+            "url": "https://example.com/page/2",
+            "status": "200",
+            "canonical_url": "https://example.com/page/2",
+            "page_analysis": json.dumps(
+                {"pagination": {"rel_prev": "https://example.com/page/1", "rel_next": None, "amphtml": "https://example.com/amp/page/3"}}
+            ),
+        },
+    ])
+    issues = pagination_issues(df)
+    assert any("amphtml" in i["message"] for i in issues)
+
+
+def test_spell_html_wayback_and_axe(monkeypatch):
+    df = pd.DataFrame([
+        {
+            "url": "https://example.com/typo",
+            "status": "200",
+            "content_excerpt": "This sentense has many misspelled wrds that should trigger heuristics.",
+            "html": LONG_HTML,
+            "page_analysis": json.dumps({"axe_violations": [{"id": "label", "description": "Missing label", "help": "Add label"}]}),
+        },
+        {"url": "https://example.com/missing", "status": "404", "page_analysis": "{}"},
+    ])
+
+    fake_spell = MagicMock()
+    fake_spell.unknown.return_value = {"sentense", "wrds", "misspelled", "heuristics"}
+    monkeypatch.setitem(__import__("sys").modules, "spellchecker", MagicMock(SpellChecker=lambda: fake_spell))
+
+    spell, _ = spell_check_issues(df, max_pages=5)
+    assert spell
+
+    html_issues, _ = html_validation_issues(df)
+    assert any("multiple title" in i["message"] for i in html_issues)
+
+    monkeypatch.setattr(
+        "website_profiling.reporting.optional_audits.requests.get",
+        lambda *a, **k: MagicMock(
+            json=lambda: {"archived_snapshots": {"closest": {"available": True, "timestamp": "20200101"}}}
+        ),
+    )
+    wb = wayback_issues(df)
+    assert wb
+
+    axe = axe_issues_from_df(df)
+    assert axe and "axe:" in axe[0]["message"]
+
+
+def test_apply_optional_audits_all_flags(monkeypatch):
+    categories = [
+        {"id": "technical_seo", "name": "Technical", "issues": [], "recommendations": []},
+        {"id": "intelligence", "name": "Content", "issues": [], "recommendations": []},
+        {"id": "html_accessibility", "name": "A11y", "issues": [], "recommendations": []},
+    ]
+    df = pd.DataFrame([
+        {
+            "url": "https://example.com/page/2",
+            "status": "200",
+            "canonical_url": "",
+            "content_type": "text/html",
+            "content_excerpt": "This sentense has many misspelled wrds that should trigger heuristics.",
+            "html": LONG_HTML,
+            "page_analysis": json.dumps(
+                {
+                    "pagination": {"rel_prev": "https://example.com/page/1", "rel_next": None},
+                    "axe_violations": [{"id": "label", "description": "Missing label", "help": "Add label"}],
+                }
+            ),
+        },
+        {"url": "https://example.com/missing", "status": "404", "page_analysis": "{}"},
+    ])
+
+    fake_spell = MagicMock()
+    fake_spell.unknown.return_value = {"sentense", "wrds", "misspelled", "heuristics"}
+    monkeypatch.setitem(__import__("sys").modules, "spellchecker", MagicMock(SpellChecker=lambda: fake_spell))
+    monkeypatch.setattr(
+        "website_profiling.reporting.optional_audits.requests.get",
+        lambda *a, **k: MagicMock(
+            json=lambda: {"archived_snapshots": {"closest": {"available": True, "timestamp": "20200101"}}}
+        ),
+    )
+
+    meta = apply_optional_audits(
+        categories,
+        df,
+        {
+            "enable_spell_check": "true",
+            "enable_html_validation": "true",
+            "enable_amp_audit": "true",
+            "enable_wayback_lookup": "true",
+            "enable_axe": "true",
+            "crawl_render_mode": "javascript",
+        },
+    )
+    assert meta.get("spell_check_pages", 0) >= 0
+    assert categories[0]["issues"]
+    assert categories[1]["issues"]
+    assert categories[2]["issues"]
+
+
+def test_optional_audits_empty_inputs():
+    assert pagination_issues(pd.DataFrame()) == []
+    assert pagination_issues(pd.DataFrame([{"url": "", "page_analysis": "{}"}])) == []
+    assert amp_audit_issues(pd.DataFrame()) == []
+    assert wayback_issues(pd.DataFrame()) == []
+    assert axe_issues_from_df(pd.DataFrame()) == []
+    assert axe_issues_from_df(pd.DataFrame([{"url": "https://x.com", "page_analysis": json.dumps({"axe_violations": ["bad"]})}])) == []
+
+
+def test_optional_audits_edge_branches(monkeypatch):
+    from website_profiling.reporting.optional_audits import _parse_page_analysis
+
+    assert _parse_page_analysis({"pagination": {}})["pagination"] == {}
+    assert _parse_page_analysis("{bad") == {}
+    assert _parse_page_analysis(None) == {}
+
+    pag_df = pd.DataFrame([{"url": "https://x.com", "page_analysis": {"pagination": {"rel_next": "https://x.com/2"}}}])
+    assert pagination_issues(pag_df) == []
+
+    dup_html = LONG_HTML + '<div id="a"></div><div id="a"></div>'
+    dup_issues, _ = html_validation_issues(pd.DataFrame([{"url": "https://x.com", "html": dup_html}]))
+    assert any("duplicate id" in i["message"] for i in dup_issues)
+
+    monkeypatch.setattr(
+        "website_profiling.reporting.optional_audits.requests.get",
+        lambda *a, **k: (_ for _ in ()).throw(RuntimeError("network")),
+    )
+    assert wayback_issues(pd.DataFrame([{"url": "https://x.com/missing", "status": "404"}])) == []
+
+
+def test_optional_audits_remaining_branches(monkeypatch):
+    assert html_validation_issues(pd.DataFrame([{"url": "https://x.com", "html": LONG_HTML}]), max_pages=0)[0] == []
+    assert wayback_issues(pd.DataFrame([{"url": "https://x.com/missing", "status": "404"}]), max_lookups=0) == []
+
+    spell_df = pd.DataFrame([
+        {"url": "https://x.com/404", "status": "404", "content_excerpt": "long enough excerpt text here for spell checker path testing"},
+        {"url": "https://x.com/short", "status": "200", "content_excerpt": "tiny"},
+        {"url": "https://x.com/empty", "status": "200", "content_excerpt": "---- ---- ---- ---- ---- ---- ---- ---- ---- ----"},
+    ])
+    fake_spell = MagicMock()
+    fake_spell.unknown.return_value = set()
+    monkeypatch.setitem(__import__("sys").modules, "spellchecker", MagicMock(SpellChecker=lambda: fake_spell))
+    assert spell_check_issues(spell_df)[0] == []
+    assert spell_check_issues(spell_df, max_pages=0)[0] == []
+
+    unclosed_html = "<html><head><title>Only</title></head><body>" + ("y" * 120)
+    html_issues, _ = html_validation_issues(pd.DataFrame([{"url": "https://x.com", "html": unclosed_html}]))
+    assert any("missing closing html" in i["message"] for i in html_issues)
+
+    amp_ok = pd.DataFrame([
+        {"url": "https://example.com/amp/article", "status": "200", "canonical_url": "https://example.com/article", "page_analysis": "{}"},
+    ])
+    assert amp_audit_issues(amp_ok) == []
+
+    categories = [{"id": "technical_seo", "name": "Technical", "issues": [], "recommendations": []}]
+    meta = apply_optional_audits(categories, amp_ok, {"enable_amp_audit": "true", "enable_html_validation": "true"})
+    assert meta.get("html_validation_parser") in ("regex", "html5lib")
+    assert "html_validation_pages" not in meta
+
+    no_a11y = [{"id": "technical_seo", "name": "Technical", "issues": [], "recommendations": []}]
+    df = pd.DataFrame([{"url": "https://x.com", "status": "200", "page_analysis": "{}"}])
+    assert apply_optional_audits(no_a11y, df, {"enable_axe": "true"}) == {}
+
+    from website_profiling.reporting.optional_audits import _accessibility_category, _technical_category
+
+    assert _accessibility_category([{"id": "technical_seo"}]) is None
+    assert _technical_category([{"id": "other"}]) is None
+
+    amp_df = pd.DataFrame(
+        [{"url": "https://example.com/amp/article", "status": "200", "canonical_url": "", "page_analysis": "{}"}]
+    )
+    amp_categories = [{"id": "technical_seo", "name": "Technical", "issues": [], "recommendations": []}]
+    apply_optional_audits(amp_categories, amp_df, {"enable_amp_audit": "true"})
+    assert amp_categories[0]["issues"]
+
+    assert wayback_issues(pd.DataFrame([{"url": "", "status": "404"}])) == []
+
+
+def test_html_validation_html5lib_parser_path(monkeypatch):
+    fake_parser = MagicMock()
+    fake_parser.parse.side_effect = ValueError("parse fail")
+    fake_html5lib = MagicMock()
+    fake_html5lib.HTMLParser.return_value = fake_parser
+    monkeypatch.setitem(__import__("sys").modules, "html5lib", fake_html5lib)
+
+    html_issues, use_parser = html_validation_issues(
+        pd.DataFrame([{"url": "https://x.com", "html": LONG_HTML}])
+    )
+    assert use_parser is True
+    assert any("parser error" in i["message"] for i in html_issues)
+
+
+def test_amp_canonical_mismatch_and_axe_static_skip():
+    amp_df = pd.DataFrame([
+        {
+            "url": "https://example.com/amp/article",
+            "status": "200",
+            "canonical_url": "https://example.com/wrong",
+            "page_analysis": json.dumps(
+                {"pagination": {"amphtml": "https://example.com/article"}}
+            ),
+        }
+    ])
+    amp_issues = amp_audit_issues(amp_df)
+    assert any("does not match" in i["message"] for i in amp_issues)
+
+    categories = [
+        {"id": "html_accessibility", "name": "A11y", "issues": [], "recommendations": []},
+    ]
+    meta = apply_optional_audits(
+        categories,
+        pd.DataFrame([{"url": "https://x.com", "status": "200", "page_analysis": "{}"}]),
+        {"enable_axe": "true", "crawl_render_mode": "static"},
+    )
+    assert meta.get("axe_skipped")
diff --git a/tests/test_rich_results.py b/tests/test_rich_results.py
new file mode 100644
index 00000000..653a7e95
--- /dev/null
+++ b/tests/test_rich_results.py
@@ -0,0 +1,71 @@
+"""Tests for Rich Results validation helpers."""
+from __future__ import annotations
+
+from unittest.mock import MagicMock, patch
+
+from website_profiling.integrations.google.rich_results import (
+    summarize_rich_results,
+    validate_urls,
+)
+
+
+def test_validate_urls_local_pass_with_schema_types():
+    links = {
+        "https://ex.com/": {
+            "url": "https://ex.com/",
+            "has_schema": True,
+            "page_analysis": {"json_ld_types": ["Organization", "WebSite"]},
+        }
+    }
+    rows = validate_urls(["https://ex.com/"], links_by_url=links)
+    assert len(rows) == 1
+    assert rows[0]["status"] == "pass"
+    assert "Organization" in rows[0]["message"]
+    assert rows[0]["provenance"] == "Crawl analysis"
+
+
+def test_validate_urls_local_warning_for_json_ld_issue():
+    links = {
+        "https://ex.com/bad": {
+            "url": "https://ex.com/bad",
+            "has_schema": True,
+            "page_analysis": {
+                "warnings": [
+                    {"code": "json_ld_missing_type", "message": "JSON-LD missing @type"},
+                ]
+            },
+        }
+    }
+    rows = validate_urls(["https://ex.com/bad"], links_by_url=links)
+    assert rows[0]["status"] == "warning"
+    assert rows[0]["issues"]
+
+
+def test_validate_urls_local_no_schema():
+    rows = validate_urls(["https://ex.com/plain"], links_by_url={})
+    assert rows[0]["status"] in ("info", "skipped")
+
+
+def test_summarize_rich_results_provenance():
+    rows = [
+        {"source": "gsc", "provenance": "Google Search Console"},
+        {"source": "crawl", "provenance": "Crawl analysis"},
+        {"source": "api", "provenance": "Google Rich Results API"},
+    ]
+    meta = summarize_rich_results(rows)
+    assert meta["checked"] == 3
+    assert meta["gsc_count"] == 1
+    assert meta["api_count"] == 1
+    assert meta["heuristic_count"] == 1
+
+
+def test_validate_urls_uses_api_when_no_gsc():
+    mock_resp = MagicMock()
+    mock_resp.status_code = 200
+    mock_resp.json.return_value = {
+        "richResultsOutcome": {"verdict": "PASS", "detectedItems": [{"richResultType": "Product"}]},
+    }
+    with patch("website_profiling.integrations.google.rich_results.requests.post", return_value=mock_resp):
+        rows = validate_urls(["https://ex.com/p"], api_key="test-key", links_by_url={})
+    assert rows[0]["source"] == "api"
+    assert rows[0]["status"] == "pass"
diff --git a/tests/test_tools_branch_coverage.py b/tests/test_tools_branch_coverage.py
new file mode 100644
index 00000000..42a8e698
--- /dev/null
+++ b/tests/test_tools_branch_coverage.py
@@ -0,0 +1,778 @@
+"""Branch and edge-path tests for website_profiling.tools (tools coverage gate)."""
+from __future__ import annotations
+
+import json
+import zipfile
+import io
+from datetime import datetime, timezone
+from unittest.mock import MagicMock, patch
+
+import pandas as pd
+import pytest
+
+from website_profiling.tools import export_artifacts
+from website_profiling.tools.audit_tools import _slice, dispatch_tool
+from website_profiling.tools.audit_tools.context import AuditToolContext as Ctx
+from website_profiling.tools.export_crawl_workbook import build_crawl_workbook_zip
+
+
+@pytest.fixture
+def ctx() -> Ctx:
+    return Ctx(property_id=1, report_id=1)
+
+
+@pytest.fixture
+def conn() -> MagicMock:
+    return MagicMock()
+
+
+def test_slice_and_context_remaining_branches() -> None:
+    assert _slice.payload_dict_slice({"x": {"a": 1, "b": 2}}, "x", fields=["a"])["data"] == {"a": 1}
+    row = {"page_analysis": {"json_ld_types": 42}}
+    assert _slice._row_schema_types_list(row) == []
+    assert _slice._parse_page_analysis({"page_analysis": {"k": 1}})["k"] == 1
+
+    conn = MagicMock()
+    c = Ctx(property_id=None, report_id=1)
+    with patch.object(Ctx, "load_payload", return_value={"canonical_domain": "Example.COM"}):
+        assert c.resolve_property_domain(conn) == "example.com"
+
+
+def test_crawl_remaining_branches(conn: MagicMock, ctx: Ctx) -> None:
+    from website_profiling.tools.audit_tools import crawl as crawl_mod
+
+    with patch.object(Ctx, "load_payload", return_value={"crawl_run_id": "bad"}), patch(
+        "website_profiling.db.crawl_store.read_edges",
+        return_value=[("https://ex.com", "https://ex.com/a")],
+    ):
+        out = crawl_mod.get_internal_links(conn, ctx, {"url": "https://ex.com"})
+        assert out["outlink_count"] == 1
+
+    df = pd.DataFrame([
+        {
+            "url": "https://ex.com/a",
+            "status": "200",
+            "title": "",
+            "noindex": "true",
+            "fetch_method": "rendered",
+            "word_count": "bad",
+            "page_analysis": json.dumps({"pagination": {"rel_next": "https://ex.com/b"}}),
+        },
+        {
+            "url": "https://ex.com/b",
+            "status": "404",
+            "title": "Has title",
+            "word_count": 5,
+            "page_analysis": json.dumps({"console_errors": "single-error"}),
+        },
+    ])
+    with patch.object(Ctx, "load_crawl_df", return_value=df):
+        assert crawl_mod.list_status_4xx_pages(conn, ctx, {})["total"] == 1
+        assert crawl_mod.get_page_analysis(conn, ctx, {"url": "https://ex.com/missing"})["error"]
+        adv = crawl_mod.search_pages_advanced(
+            conn,
+            ctx,
+            {
+                "noindex_only": "true",
+                "missing_title": "true",
+                "has_pagination": "true",
+                "fetch_method": "rendered",
+            },
+        )
+        assert adv["total"] == 1
+        assert crawl_mod.search_pages_advanced(
+            conn, ctx, {"min_word_count": "x", "word_count": 10}
+        )["total"] >= 0
+        assert crawl_mod.search_pages_advanced(
+            conn, ctx, {"max_word_count": "y", "word_count": 10}
+        )["total"] >= 0
+        assert crawl_mod.search_pages_advanced(conn, ctx, {"status": "200", "url_contains": "ex"})["total"] == 1
+        assert crawl_mod.list_pages_with_console_errors(conn, ctx, {})["total"] == 1
+
+    with patch.object(Ctx, "load_crawl_df", return_value=pd.DataFrame()):
+        assert crawl_mod._status_prefix_pages(conn, ctx, {}, "4")["total"] == 0
+        assert crawl_mod.get_page_analysis(conn, ctx, {"url": "https://ex.com"})["error"]
+        assert crawl_mod.search_pages_advanced(conn, ctx, {})["total"] == 0
+        assert crawl_mod.list_pages_with_console_errors(conn, ctx, {})["total"] == 0
+
+    payload = {
+        "links": "bad",
+        "graph_edges": "bad",
+        "graph_nodes": ["a"],
+    }
+    with patch.object(Ctx, "load_payload", return_value=payload):
+        links = crawl_mod.get_crawl_links_table(conn, ctx, {"url_contains": "ex"})
+        assert links["links"] == []
+        edges = crawl_mod.get_graph_edges_sample(conn, ctx, {})
+        assert edges["graph_node_count"] == 1
+
+
+def test_crawl_lists_remaining_branches(conn: MagicMock, ctx: Ctx) -> None:
+    from website_profiling.tools.audit_tools import crawl_lists as cl_mod
+
+    assert cl_mod._is_2xx("") is False
+    df_no_status = pd.DataFrame([{"url": "https://ex.com"}])
+    assert len(cl_mod._success_df(df_no_status)) == 1
+
+    with patch.object(Ctx, "load_crawl_df", return_value=pd.DataFrame()):
+        assert cl_mod._filter_crawl_pages(conn, ctx, {}, predicate=lambda r: True, projection=lambda r: r)["total"] == 0
+
+    with patch.object(Ctx, "load_payload", return_value={"content_urls": {"missing_alt": "bad"}}):
+        out = cl_mod._content_urls_list(conn, ctx, {}, "missing_alt")
+        assert out["total"] == 0
+
+    df = pd.DataFrame([
+        {
+            "url": "https://ex.com/a",
+            "status": "200",
+            "images_without_alt": "bad",
+            "heading_sequence": "",
+            "redirect_chain_length": "bad",
+        },
+    ])
+    with patch.object(Ctx, "load_crawl_df", return_value=df):
+        assert cl_mod.list_pages_with_missing_alt(conn, ctx, {})["total"] == 0
+        assert cl_mod.list_pages_skipped_headings(conn, ctx, {})["total"] == 0
+        assert cl_mod.list_long_redirect_chains(conn, ctx, {})["total"] == 0
+
+    with patch.object(Ctx, "load_crawl_df", return_value=pd.DataFrame([{"url": "https://ex.com", "status": "200"}])):
+        assert cl_mod.list_pages_missing_viewport(conn, ctx, {})["note"]
+
+    payload = {"social_coverage": {"og_image_missing": ["https://ex.com/no-og"]}}
+    with patch.object(Ctx, "load_payload", return_value=payload), patch.object(
+        Ctx, "load_crawl_df", return_value=pd.DataFrame([{"url": "https://ex.com", "status": "200", "og_image": ""}]),
+    ):
+        assert cl_mod.list_pages_missing_og_image(conn, ctx, {})["total"] == 1
+
+    with patch.object(Ctx, "load_payload", return_value={}):
+        assert cl_mod.get_top_pages_by_pagerank(conn, ctx, {})["error"]
+
+    bad_pr = {"top_pages": ["bad", {"url": "https://ex.com", "pagerank": "n/a"}]}
+    with patch.object(Ctx, "load_payload", return_value=bad_pr):
+        assert cl_mod.get_top_pages_by_pagerank(conn, ctx, {})["total"] == 0
+
+
+def test_backlinks_charts_compare_and_content(conn: MagicMock, ctx: Ctx) -> None:
+    from website_profiling.tools.audit_tools import backlinks as bl_mod
+    from website_profiling.tools.audit_tools import charts as charts_mod
+    from website_profiling.tools.audit_tools import compare_slices as cmp_mod
+    from website_profiling.tools.audit_tools import content as content_mod
+
+    assert bl_mod.get_gsc_links_import_status(conn, Ctx(property_id=None), {})["error"]
+    assert bl_mod.get_gsc_sample_links(conn, Ctx(property_id=None), {})["error"]
+    assert bl_mod.get_gsc_latest_links(conn, Ctx(property_id=None), {})["error"]
+    assert bl_mod.get_third_party_links_overlay(conn, Ctx(property_id=None), {})["error"]
+    assert bl_mod.get_backlinks_velocity(conn, Ctx(property_id=None), {})["error"]
+
+    with patch.object(Ctx, "load_payload", return_value={}):
+        assert charts_mod.get_outlink_distribution(conn, ctx, {})["error"]
+
+    with patch("website_profiling.tools.audit_tools.report.get_report_summary", return_value={"error": "x"}):
+        assert charts_mod.get_issue_priority_breakdown(conn, ctx, {})["error"]
+    with patch(
+        "website_profiling.tools.audit_tools.report.get_report_summary",
+        return_value={"issue_counts": "bad", "total_issues": 0},
+    ):
+        assert charts_mod.get_issue_priority_breakdown(conn, ctx, {})["items"] == []
+
+    err = {"error": "baseline missing"}
+    with patch("website_profiling.tools.audit_tools.compare_slices.load_compare_pair", return_value=(None, None, None, None, err)):
+        for fn in (
+            cmp_mod.compare_security_deltas,
+            cmp_mod.compare_content_metrics,
+            cmp_mod.compare_google_metrics,
+            cmp_mod.compare_priority_counts,
+            cmp_mod.compare_health_score_delta,
+        ):
+            assert fn(conn, ctx, {"baseline_report_id": 1})["error"] == err["error"]
+
+    clusters = ["bad", {"representative_url": "https://ex.com/a", "member_urls": []}]
+    with patch.object(Ctx, "load_payload", return_value={"content_duplicates": clusters}):
+        assert content_mod.get_duplicate_cluster(conn, ctx, {"cluster_index": 0})["error"]
+
+
+def test_report_report_extras_keywords_ops(conn: MagicMock, ctx: Ctx) -> None:
+    from website_profiling.tools.audit_tools import report as report_mod
+    from website_profiling.tools.audit_tools import report_extras as rex_mod
+    from website_profiling.tools.audit_tools import keywords as kw_mod
+    from website_profiling.tools.audit_tools import ops as ops_mod
+
+    assert report_mod._normalize_priority("weird") == "weird"
+    with patch.object(Ctx, "load_payload", return_value={}):
+        assert report_mod.list_issues(conn, ctx, {"limit": "bad"})["error"]
+        assert report_mod.search_issues(conn, ctx, {"limit": "bad"})["error"]
+    with patch.object(Ctx, "load_payload", return_value={"executive_summary": None}):
+        assert report_mod.get_executive_summary(conn, ctx, {})["missing"]
+        assert report_mod.get_report_meta(conn, ctx, {})["missing"]
+        assert report_mod.get_site_level(conn, ctx, {})["missing"]
+
+    payload = {
+        "categories": [
+            {
+                "id": "x",
+                "issues": [{"priority": "High", "message": "m", "url": "https://ex.com", "llm_recommendation": "fix"}],
+                "recommendations": "bad",
+            },
+            "bad-cat",
+        ],
+        "recommendations": "bad",
+        "ml_errors": "bad",
+    }
+    with patch.object(Ctx, "load_payload", return_value=payload):
+        issues = report_mod.list_issues(conn, ctx, {"sort": "impact", "priority": "high", "url_contains": "ex"})
+        assert issues["total"] == 1
+        searched = report_mod.search_issues(conn, ctx, {"message_contains": "m"})
+        assert searched["total"] == 1
+        assert rex_mod.get_audit_recommendations(conn, ctx, {})["count"] == 0
+        assert rex_mod.get_ml_errors(conn, ctx, {})["count"] == 0
+        assert rex_mod.list_audit_categories(conn, ctx, {})["count"] == 1
+        assert rex_mod.list_issues_with_ai_fixes(conn, ctx, {})["total"] == 1
+
+    assert rex_mod.get_category_recommendations(conn, ctx, {})["error"]
+    with patch.object(Ctx, "load_payload", return_value={"categories": [{"id": "y", "recommendations": ["r"]}]}):
+        assert rex_mod.get_category_recommendations(conn, ctx, {"category_id": "y"})["recommendations"] == ["r"]
+
+    assert kw_mod.get_keyword_history(conn, Ctx(property_id=None), {"keyword": "x"})["error"]
+    assert kw_mod.get_keyword_history(conn, ctx, {})["error"]
+    assert kw_mod.get_keyword_serp_overlay(conn, Ctx(property_id=None), {})["error"]
+    assert kw_mod.list_keywords_by_action(conn, ctx, {})["error"]
+    with patch.object(Ctx, "load_keywords", return_value={"rows": [{"keyword": "a", "gsc_position": 5}]}):
+        ranged = kw_mod.list_keywords_by_position(conn, ctx, {"min_position": 1, "max_position": 10})
+        assert ranged["total"] == 1
+        assert kw_mod.list_keywords_by_position(conn, ctx, {"min_position": 10, "max_position": 1})["total"] == 0
+        assert kw_mod.list_keywords_by_impressions(conn, ctx, {"min_impressions": 5})["total"] == 0
+
+    with patch.object(Ctx, "load_keywords", return_value={"rows": [{"keyword": "a", "serp_estimated_competition": 0.5}]}):
+        assert kw_mod.get_keyword_serp_overlay(conn, ctx, {})["total"] == 1
+
+    log_row = {
+        "upload_id": 1,
+        "filename": "access.log",
+        "analysis": {},
+        "line_count": 1,
+    }
+    with patch("website_profiling.tools.audit_tools.ops._load_log_analysis", return_value=log_row):
+        assert ops_mod.get_log_top_paths(conn, ctx, {})["total"] == 0
+        assert ops_mod.list_log_only_paths(conn, ctx, {})["total"] == 0
+        assert ops_mod.list_crawl_only_paths(conn, ctx, {})["total"] == 0
+        assert ops_mod.get_log_googlebot_stats(conn, ctx, {})["parsed_lines"] == 0
+
+
+def test_lighthouse_links_google_health_security(conn: MagicMock, ctx: Ctx) -> None:
+    from website_profiling.tools.audit_tools import lighthouse as lh_mod
+    from website_profiling.tools.audit_tools import links as links_mod
+    from website_profiling.tools.audit_tools import google as google_mod
+    from website_profiling.tools.audit_tools import health as health_mod
+    from website_profiling.tools.audit_tools import security as sec_mod
+    from website_profiling.tools.audit_tools import indexation_tools as idx_mod
+
+    with patch.object(Ctx, "load_payload", return_value={"lighthouse_by_url": "bad"}), patch(
+        "website_profiling.tools.audit_tools.lighthouse.read_lighthouse_page_summaries",
+        return_value={"https://ex.com": {"performance": 10}},
+    ):
+        assert lh_mod.get_lighthouse_for_url(conn, ctx, {"url": "https://ex.com"})["lighthouse"]
+
+    lh_payload = {
+        "lighthouse_by_url": {
+            "https://ex.com/slow": {"scores": {"performance": 20}},
+            "https://ex.com/seo": {"scores": {"seo": 40}},
+            "https://ex.com/a11y": {
+                "scores": {"accessibility": "bad"},
+                "category_scores": {"accessibility_score": 30},
+            },
+            "https://ex.com/bp": {
+                "median_metrics": {"best_practices": 30},
+            },
+            "https://ex.com/cwv": {
+                "median_metrics": {"lcp_ms": "bad", "cls": "bad", "tbt_ms": "bad"},
+            },
+        },
+    }
+    with patch.object(Ctx, "load_payload", return_value=lh_payload):
+        assert lh_mod.list_slow_pages(conn, ctx, {})["total"] >= 1
+        assert lh_mod.list_lighthouse_poor_seo_pages(conn, ctx, {})["total"] >= 1
+        assert lh_mod.list_lighthouse_poor_accessibility_pages(conn, ctx, {})["total"] >= 1
+        assert lh_mod.list_lighthouse_poor_best_practices_pages(conn, ctx, {})["total"] >= 1
+        assert lh_mod.list_lighthouse_cwv_failures(conn, ctx, {})["total"] == 0
+        assert lh_mod.get_crux_summary(conn, ctx, {})["missing"]
+
+    with patch.object(Ctx, "load_payload", return_value={"issues": {"broken": []}}):
+        assert links_mod.list_broken_link_sources(conn, ctx, {})["total"] == 0
+        assert links_mod.get_link_rel_summary(conn, ctx, {})["total_edges"] == 0
+        assert links_mod.get_inlink_anchors(conn, ctx, {"url": "https://ex.com"})["total"] == 0
+        assert links_mod.list_nofollow_internal_links(conn, ctx, {})["total"] == 0
+
+    with patch.object(Ctx, "load_google", return_value={"ga4": {"top_pages": [{"path": "/home", "sessions": 3}]}}):
+        assert google_mod.get_ga4_page_metrics(conn, ctx, {"path": "/home"})["metrics"]["sessions"] == 3
+
+    with patch.object(Ctx, "load_payload", return_value={"indexation_coverage": {}}):
+        assert idx_mod.list_indexation_gaps(conn, ctx, {"gap_type": "not-real"})["error"]
+
+    conn.execute = MagicMock(return_value=MagicMock(fetchall=MagicMock(return_value=[])))
+    with patch.object(Ctx, "load_payload", return_value={"categories": []}):
+        assert health_mod.get_health_history(conn, ctx, {})["count"] == 0
+
+    with patch.object(Ctx, "load_payload", return_value={"security_findings": [{"severity": "High", "finding_type": "x"}]}):
+        assert sec_mod.list_security_findings_by_type(conn, ctx, {"finding_type": "x"})["total"] == 1
+
+
+def test_image_tools_and_misc_audit_modules(conn: MagicMock, ctx: Ctx) -> None:
+    from website_profiling.tools.audit_tools import image_tools as img_mod
+    from website_profiling.tools.audit_tools import onpage as onpage_mod
+    from website_profiling.tools.audit_tools import llm_tools as llm_mod
+    from website_profiling.tools.audit_tools import workflow as wf_mod
+
+    payload = {
+        "lighthouse_diagnostics": ["bad", {"lighthouse_audit_id": "uses-optimized-images", "title": "Images"}] * 10,
+        "links": [
+            "bad",
+            {
+                "url": "https://ex.com",
+                "og_image": "https://cdn.ex.com/og.png",
+                "twitter_image": "https://cdn.ex.com/tw.png",
+                "page_analysis": {"image_urls": ["https://cdn.ex.com/hero.jpg", "https://cdn.ex.com/hero.jpg"]},
+            },
+        ],
+        "image_inventory": [
+            {"url": "https://cdn.ex.com/big.png", "size_bytes": 500_000, "content_type": "image/png"},
+            {"url": "https://cdn.ex.com/modern.webp", "size_bytes": 500_000, "content_type": "image/webp"},
+            {"url": "https://cdn.ex.com/unknown", "size_bytes": 500_000, "content_type": ""},
+        ],
+        "image_inventory_summary": {"unoptimized_min_kb": 200},
+        "content_urls": {
+            "missing_alt": [{"url": "https://ex.com/a"}],
+            "missing_lazy": [{"url": "https://ex.com/l"}],
+            "missing_dimensions": [{"url": "https://ex.com/d"}],
+        },
+        "social_coverage": {"og_image_missing": ["https://ex.com/no-og"]},
+    }
+    df = pd.DataFrame([
+        {
+            "url": "https://ex.com/a",
+            "status": "200",
+            "images_without_alt": 1,
+            "img_without_lazy": 1,
+            "img_without_dimensions": 1,
+            "images_total": 3,
+        },
+    ])
+    with patch.object(Ctx, "load_payload", return_value=payload), patch.object(Ctx, "load_crawl_df", return_value=df):
+        assert img_mod.list_site_image_urls(conn, ctx, {"kind": "twitter"})["total"] == 1
+        assert img_mod.list_largest_images(conn, ctx, {"min_size_kb": "bad"})["inventory_available"] is True
+        assert img_mod.list_unoptimized_images(conn, ctx, {})["total"] >= 1
+        assert img_mod.list_pages_without_lazy_images(conn, ctx, {})["total"] >= 1
+        assert img_mod.list_pages_with_images_missing_dimensions(conn, ctx, {})["total"] >= 1
+
+    with patch.object(Ctx, "load_payload", return_value={"issues": {"seo": [{"message": "m"}]}}):
+        assert onpage_mod.list_seo_onpage_issues(conn, ctx, {})["total"] == 1
+
+    with patch.object(Ctx, "load_crawl_df", return_value=pd.DataFrame([{"url": "https://ex.com", "status": "200", "noindex": "true"}])):
+        assert onpage_mod.list_pages_noindex(conn, ctx, {})["total"] == 1
+
+    with patch("website_profiling.tools.audit_tools.llm_tools.run_page_coach", return_value={"coach": "ok"}):
+        assert llm_mod.get_page_coach(conn, ctx, {"url": "https://ex.com"})["coach"] == "ok"
+
+    assert wf_mod.list_issue_workflow(conn, Ctx(property_id=None), {})["error"]
+
+
+def test_export_artifacts_workbook_and_custom(tmp_path, monkeypatch, conn: MagicMock, ctx: Ctx) -> None:
+    monkeypatch.setenv("DATA_DIR", str(tmp_path))
+    artifact = export_artifacts.save_artifact(b"data", filename="x.bin", mime_type="application/octet-stream")
+    aid = artifact["artifact_id"]
+    meta_path = tmp_path / "exports" / f"{aid}.meta.json"
+    data_path = tmp_path / "exports" / f"{aid}.bin"
+    assert data_path.exists()
+    data_path.unlink()
+    assert export_artifacts.read_artifact_bytes(aid) is None
+    export_artifacts.delete_artifact(aid)
+    assert not meta_path.exists()
+
+    spec_id = export_artifacts.save_report_spec({"title": "T"})
+    assert export_artifacts.read_report_spec(spec_id)["title"] == "T"
+    assert export_artifacts.read_report_spec("not-a-uuid") is None
+
+    from website_profiling.tools import export_crawl_workbook as wb_mod
+
+    assert wb_mod._parse_custom_fields({"price": 9.99}) == {"price": "9.99"}
+    assert wb_mod._parse_custom_fields("{bad") == {}
+    assert wb_mod._parse_custom_fields("[]") == {}
+
+    raw = build_crawl_workbook_zip({
+        "links": [{"url": "https://ex.com/p", "custom_fields": '{"price":"9.99"}'}],
+        "categories": ["bad", {"name": "SEO", "issues": ["bad", {"message": "x", "priority": "Low"}]}],
+    })
+    with zipfile.ZipFile(io.BytesIO(raw)) as zf:
+        assert "custom_fields.csv" in zf.namelist()
+
+    from website_profiling.tools.export_custom import render_custom_report_pdf
+
+    try:
+        render_custom_report_pdf("T", {"site_name": "Ex"}, [], [])
+    except Exception:
+        pass
+
+
+def test_tools_remaining_branch_coverage(conn: MagicMock, ctx: Ctx, tmp_path, monkeypatch) -> None:
+    from website_profiling.tools.audit_tools import backlinks as bl_mod
+    from website_profiling.tools.audit_tools import charts as charts_mod
+    from website_profiling.tools.audit_tools import content as content_mod
+    from website_profiling.tools.audit_tools import crawl as crawl_mod
+    from website_profiling.tools.audit_tools import crawl_lists as cl_mod
+    from website_profiling.tools.audit_tools import export_tools as et_mod
+    from website_profiling.tools.audit_tools import google as google_mod
+    from website_profiling.tools.audit_tools import health as health_mod
+    from website_profiling.tools.audit_tools import image_tools as img_mod
+    from website_profiling.tools.audit_tools import issues as issues_mod
+    from website_profiling.tools.audit_tools import keywords as kw_mod
+    from website_profiling.tools.audit_tools import lighthouse as lh_mod
+    from website_profiling.tools.audit_tools import links as links_mod
+    from website_profiling.tools.audit_tools import llm_tools as llm_mod
+    from website_profiling.tools.audit_tools import onpage as onpage_mod
+    from website_profiling.tools.audit_tools import ops as ops_mod
+    from website_profiling.tools.audit_tools import report as report_mod
+    from website_profiling.tools.audit_tools import report_extras as rex_mod
+    from website_profiling.tools.audit_tools import security as sec_mod
+    from website_profiling.tools import export_crawl_workbook as wb_mod
+    from website_profiling.tools.export_custom import (
+        _section_html_tool_result,
+        resolve_section_results,
+        validate_sections,
+    )
+
+    monkeypatch.setenv("DATA_DIR", str(tmp_path))
+
+    assert cl_mod._heading_skipped("h1,h3") is True
+    assert cl_mod._heading_skipped("h1,h2") is False
+    with patch.object(Ctx, "load_payload", return_value={}), patch.object(
+        Ctx,
+        "load_crawl_df",
+        return_value=pd.DataFrame([{"url": "https://ex.com", "status": "200", "og_image": ""}]),
+    ):
+        assert cl_mod.list_pages_missing_og_image(conn, ctx, {})["total"] == 1
+    with patch.object(Ctx, "load_payload", return_value={"top_pages": "bad"}):
+        assert cl_mod.get_top_pages_by_pagerank(conn, ctx, {})["total"] == 0
+
+    with patch.object(Ctx, "load_gsc_links", return_value={"sample_links": [{"url": "https://ex.com"}]}):
+        assert bl_mod.get_gsc_sample_links(conn, ctx, {})["total"] == 1
+    with patch.object(Ctx, "load_gsc_links", return_value=None):
+        assert bl_mod.get_gsc_sample_links(conn, ctx, {})["missing"]
+
+    with patch.object(Ctx, "load_payload", return_value={"outlink_labels": ["a"], "outlink_counts": [1]}):
+        assert charts_mod.get_outlink_distribution(conn, ctx, {})["items"]
+
+    with patch.object(Ctx, "load_payload", return_value={}):
+        assert content_mod.get_duplicate_cluster(conn, ctx, {})["error"]
+    with patch.object(Ctx, "load_payload", return_value={"content_duplicates": "bad"}):
+        assert content_mod.get_duplicate_cluster(conn, ctx, {})["missing"]
+    clusters = [{"representative_url": "https://ex.com/a", "member_urls": []}, "bad"]
+    with patch.object(Ctx, "load_payload", return_value={"content_duplicates": clusters}):
+        assert content_mod.get_duplicate_cluster(conn, ctx, {"url": "https://ex.com/a"})["cluster_index"] == 0
+
+    no_pag_df = pd.DataFrame([{
+        "url": "https://ex.com/c",
+        "status": "200",
+        "noindex": "true",
+        "title": "",
+        "fetch_method": "rendered",
+        "page_analysis": json.dumps({"pagination": {}}),
+    }])
+    with patch.object(Ctx, "load_crawl_df", return_value=no_pag_df):
+        assert crawl_mod.search_pages_advanced(conn, ctx, {"has_pagination": "true"})["total"] == 0
+    crawl_df = pd.DataFrame([{"url": "https://ex.com/a", "status": "200", "title": "t"}])
+    with patch.object(Ctx, "load_crawl_df", return_value=crawl_df):
+        assert crawl_mod.search_pages_advanced(conn, ctx, {"url_contains": "nomatch"})["total"] == 0
+
+    assert img_mod._preview_from_bucket(["https://ex.com/s"])["total"] == 1
+    assert img_mod._preview_from_bucket([{"bad": 1}])["total"] == 1
+    assert img_mod._int_val("bad") == 0
+    assert img_mod._threshold_kb({"min_size_kb": "bad"}, {"image_inventory_summary": {"unoptimized_min_kb": "bad"}}) == 200
+    with patch.object(Ctx, "load_payload", return_value={}):
+        assert img_mod.get_image_audit_summary(conn, ctx, {})["error"]
+        assert img_mod.list_site_image_urls(conn, ctx, {})["error"]
+        assert img_mod.list_lighthouse_image_opportunities(conn, ctx, {})["error"]
+        assert img_mod.list_largest_images(conn, ctx, {})["error"]
+        assert img_mod.list_unoptimized_images(conn, ctx, {})["error"]
+        assert img_mod.list_images_needing_attention(conn, ctx, {})["error"]
+    link_payload = {
+        "links": [{
+            "url": "https://ex.com",
+            "og_image": "",
+            "twitter_image": "https://cdn.ex.com/tw.png",
+            "page_analysis": {"image_urls": ["", "https://cdn.ex.com/a.jpg", "https://cdn.ex.com/a.jpg"]},
+        }],
+        "lighthouse_diagnostics": ["bad", {"lighthouse_audit_id": "uses-optimized-images", "title": "T"}] * 10,
+        "image_inventory": [
+            {"url": "https://cdn.ex.com/err.png", "size_bytes": 500_000, "content_type": "image/png", "error": "timeout"},
+            {"url": "https://cdn.ex.com/big.webp", "size_bytes": 500_000, "content_type": "image/webp", "source_pages": ["https://ex.com/a"]},
+            {"url": "https://cdn.ex.com/unk", "size_bytes": 500_000, "content_type": ""},
+        ],
+        "image_inventory_summary": {"unoptimized_min_kb": 200},
+        "content_urls": {"missing_alt": [{"url": "https://ex.com/a"}]},
+    }
+    with patch.object(Ctx, "load_payload", return_value=link_payload):
+        assert img_mod.list_site_image_urls(conn, ctx, {"kind": "content"})["total"] == 1
+        assert img_mod.list_unoptimized_images(conn, ctx, {})["total"] >= 1
+        assert img_mod.list_images_needing_attention(conn, ctx, {})["total"] >= 1
+    with patch.object(Ctx, "load_payload", return_value={"content_urls": {}}), patch.object(Ctx, "load_crawl_df", return_value=None):
+        assert img_mod.list_pages_without_lazy_images(conn, ctx, {})["total"] == 0
+
+    assert google_mod.get_ga4_summary(conn, ctx, {})["error"]
+    assert google_mod.get_gsc_page_query_slice(conn, ctx, {})["error"]
+    assert google_mod.get_ga4_page_metrics(conn, ctx, {})["error"]
+    with patch.object(Ctx, "load_google", return_value={"ga4": {"top_pages": ["bad", {"path": "/home", "sessions": 2}]}}):
+        assert google_mod.get_ga4_page_metrics(conn, ctx, {"path": "https://ex.com/home"})["metrics"]["sessions"] == 2
+        assert google_mod.get_ga4_page_metrics(conn, ctx, {"path": "/missing"})["missing"]
+
+    conn.execute = MagicMock(return_value=MagicMock(fetchall=MagicMock(return_value=[])))
+    with patch.object(Ctx, "resolve_property_domain", return_value="ex.com"), patch.object(Ctx, "load_payload", return_value={"canonical_domain": "ex.com"}):
+        c = Ctx(property_id=None, report_id=1)
+        assert health_mod.list_report_history(conn, c, {})["count"] == 0
+    assert health_mod.get_category_health_history(conn, Ctx(property_id=None), {})["error"]
+    row = MagicMock()
+    row.__getitem__ = lambda self, i: ["{}", datetime.now(timezone.utc), 1, 80][i]
+    conn.execute = MagicMock(return_value=MagicMock(fetchall=MagicMock(return_value=[row])))
+    with patch.object(Ctx, "load_payload", return_value={}):
+        hist = health_mod.get_category_health_history(conn, ctx, {"category_id": "seo", "limit": "bad"})
+        assert hist["count"] == 1
+
+    assert kw_mod.get_striking_distance_keywords(conn, Ctx(property_id=None), {})["error"]
+    with patch.object(Ctx, "load_keywords", return_value={"striking_distance": [{"keyword": "a"}]}):
+        assert kw_mod.get_striking_distance_keywords(conn, ctx, {})["total"] == 1
+    with patch.object(Ctx, "load_keywords", return_value={"rows": [{"keyword": "a", "gsc_impressions": "bad"}]}):
+        assert kw_mod.list_keywords_by_impressions(conn, ctx, {"min_impressions": "bad"})["error"]
+        assert kw_mod.list_keywords_by_impressions(conn, ctx, {"min_impressions": 0})["total"] == 1
+    with patch.object(Ctx, "load_keywords", return_value={"rows": [{"keyword": "a", "gsc_position": 5}]}):
+        assert kw_mod.list_keywords_by_position(conn, ctx, {"min_position": 1, "max_position": 10})["total"] == 1
+
+    lh_bad = {
+        "lighthouse_by_url": {
+            "bad": "x",
+            "https://ex.com/slow": {"performance": 20},
+            "https://ex.com/seo": {"scores": {"seo": 40}},
+            "https://ex.com/cwv": {"median_metrics": {"lcp_ms": 5000, "cls": 0.5, "tbt_ms": 500}},
+            "https://ex.com/a11y": {"scores": {"accessibility": "bad"}, "category_scores": {"accessibility_score": 30}},
+        },
+        "crux_summary": {"origin": "https://ex.com"},
+        "lighthouse_human_summary": "Human text",
+    }
+    with patch.object(Ctx, "load_payload", return_value=lh_bad):
+        assert lh_mod.list_slow_pages(conn, ctx, {})["total"] >= 1
+        assert lh_mod.list_lighthouse_poor_seo_pages(conn, ctx, {})["total"] >= 1
+        assert lh_mod.list_lighthouse_cwv_failures(conn, ctx, {})["total"] >= 1
+        assert lh_mod.get_crux_summary(conn, ctx, {})["data"]
+        assert lh_mod.get_lighthouse_human_summary(conn, ctx, {})["has_summary"]
+    with patch.object(Ctx, "load_payload", return_value={}):
+        assert lh_mod._list_lighthouse_poor_category(conn, ctx, {}, score_keys=("x",), result_key="x", threshold_arg="t", default_threshold=50)["error"]
+
+    with patch.object(Ctx, "load_payload", return_value={}):
+        assert links_mod.list_broken_link_sources(conn, ctx, {})["error"]
+        assert links_mod.get_link_rel_summary(conn, ctx, {})["error"]
+        assert links_mod.get_inlink_anchors(conn, ctx, {})["error"]
+        assert links_mod.list_nofollow_internal_links(conn, ctx, {})["error"]
+    broken_payload = {
+        "issues": {"broken": [{"url": "https://ex.com/broken"}]},
+        "graph_edges": [("https://ex.com", "https://ex.com/broken")],
+        "link_edges": [{"link_type": "internal", "is_nofollow": True, "from": "a", "to": "b"}],
+        "inlink_anchor_matrix": [{"target_url": "https://ex.com/t", "anchor": "x"}],
+    }
+    with patch.object(Ctx, "load_payload", return_value=broken_payload):
+        assert links_mod.list_broken_link_sources(conn, ctx, {})["total"] == 1
+        assert links_mod.get_link_rel_summary(conn, ctx, {})["total_edges"] >= 0
+        assert links_mod.get_inlink_anchors(conn, ctx, {"url": "https://ex.com/t"})["total"] == 1
+        assert links_mod.list_nofollow_internal_links(conn, ctx, {})["total"] == 1
+
+    with patch.object(Ctx, "load_payload", return_value={"categories": [{"id": "c", "name": "C", "issues": []}]}):
+        assert report_mod.get_report_summary(conn, ctx, {})["total_issues"] == 0
+        assert report_mod.search_issues(conn, ctx, {"limit": "bad", "category_id": "c", "url_contains": "ex", "message_contains": "m"})["total"] == 0
+        assert report_mod.get_critical_issues(conn, ctx, {})["total"] == 0
+    cat_payload = {
+        "categories": [
+            {"id": "c", "name": "C", "issues": [{"message": "m", "priority": "Low"}]},
+            "bad",
+        ],
+    }
+    with patch.object(Ctx, "load_payload", return_value=cat_payload):
+        assert issues_mod.get_category_issues(conn, ctx, {"category_id": "c"})["issue_count"] == 1
+        assert rex_mod.get_category_recommendations(conn, ctx, {"category_id": "missing"})["error"]
+        assert rex_mod.list_issues_with_ai_fixes(conn, ctx, {})["total"] == 0
+
+    with patch.object(Ctx, "load_payload", return_value={}):
+        assert sec_mod.get_security_findings_summary(conn, ctx, {})["error"]
+        assert sec_mod.list_security_findings_by_type(conn, ctx, {"finding_type": "x"})["error"]
+
+    with patch.object(Ctx, "load_payload", return_value={"content_urls": "bad"}):
+        assert onpage_mod._content_urls_bucket(conn, ctx, {}, "missing_h1")["missing"]
+    with patch.object(Ctx, "load_payload", return_value={"content_urls": {"missing_h1": "bad"}}):
+        assert onpage_mod._content_urls_bucket(conn, ctx, {}, "missing_h1")["total"] == 0
+
+    assert ops_mod._parse_analysis_field('{"top_paths": []}') == {"top_paths": []}
+    assert ops_mod._parse_analysis_field("{bad") == {}
+    assert ops_mod.get_log_analysis_by_id(conn, Ctx(property_id=None), {"upload_id": 1})["error"]
+    log_row = {
+        "upload_id": 2,
+        "filename": "access.log",
+        "analysis": {"top_paths": "bad", "parsed_lines": 10, "googlebot_hits": 2, "crawl_compare": {"log_only_paths": ["/a"], "crawl_only_paths": ["/b"]}},
+        "line_count": 1,
+    }
+    with patch("website_profiling.tools.audit_tools.ops._load_log_analysis", return_value=log_row):
+        assert ops_mod.get_log_top_paths(conn, ctx, {})["total"] == 0
+        assert ops_mod.list_log_only_paths(conn, ctx, {})["total"] == 1
+        assert ops_mod.list_crawl_only_paths(conn, ctx, {})["total"] == 1
+        assert ops_mod.get_log_googlebot_stats(conn, ctx, {})["googlebot_ratio"] == 0.2
+
+    props = [{"id": 1, "name": "Ex", "canonical_domain": "ex.com"}]
+    snap = MagicMock()
+    snap.__getitem__ = lambda self, i: [80, datetime.now(timezone.utc), 1, "not-json"][i]
+    conn.execute = MagicMock(return_value=MagicMock(fetchone=MagicMock(return_value=snap)))
+    with patch("website_profiling.tools.audit_tools.llm_tools.list_properties_public", return_value=props):
+        portfolio = llm_mod.get_portfolio_summary(conn, ctx, {})
+        assert portfolio["properties"][0]["issue_counts"] == {}
+
+    assert wb_mod._parse_custom_fields("   ") == {}
+    rows, cols = wb_mod._custom_field_rows([{"url": "", "custom_fields": '{"a":"1"}'}, {"custom_extract": "x"}])
+    assert rows == [] and cols
+
+    assert export_artifacts.read_report_spec("00000000-0000-0000-0000-000000000000") is None
+    aid = export_artifacts.save_artifact(b"x", filename="y.bin", mime_type="application/octet-stream")["artifact_id"]
+    with patch("website_profiling.tools.export_artifacts.os.remove", side_effect=OSError("denied")):
+        export_artifacts.delete_artifact(aid)
+
+    assert _section_html_tool_result("H", {"keywords": [{"k": "v"}]}) != ""
+    _, err = validate_sections([{"type": "weird"}])
+    assert err
+    results = resolve_section_results(conn, ctx, {}, [{"type": "weird"}], lambda *a, **k: {})
+    assert results == [None]
+
+    with patch.object(Ctx, "load_payload", return_value={}):
+        assert et_mod.compose_custom_report(conn, ctx, {"title": "T", "sections": [{"type": "executive_summary"}]})["error"]
+        assert et_mod.export_custom_report(conn, ctx, {"title": "T", "sections": [{"type": "executive_summary"}]})["error"]
+    with patch.object(Ctx, "load_payload", return_value={"site_name": "Ex"}), patch(
+        "website_profiling.tools.audit_tools.export_tools._dispatch",
+        return_value={"error": "tool failed"},
+    ):
+        assert et_mod.export_list_as_csv(conn, ctx, {"tool_name": "list_broken_links"})["error"] == "tool failed"
+    with patch.object(Ctx, "load_payload", return_value={"site_name": "Ex"}), patch(
+        "website_profiling.tools.audit_tools.export_tools._dispatch",
+        return_value={"pages": [{"url": "https://ex.com"}]},
+    ):
+        out = et_mod.export_list_as_csv(conn, ctx, {"tool_name": "list_broken_links"})
+        assert out.get("total") == 1
+    with patch.object(Ctx, "load_payload", return_value={"site_name": "Ex"}):
+        assert et_mod.compose_custom_report(conn, ctx, {"title": "T", "sections": [{"type": "nope"}]})["error"]
+        assert et_mod.export_custom_report(
+            conn,
+            ctx,
+            {"title": "T", "sections": [{"type": "tool", "tool_name": "export_audit_report", "tool_args": {}}]},
+        )["error"]
+        assert et_mod.export_custom_report(conn, ctx, {"sections": [{"type": "nope"}]})["error"]
+        bad_spec = export_artifacts.save_report_spec({"title": "Bad", "sections": [{"type": "nope"}]})
+        assert et_mod.export_custom_report(conn, ctx, {"report_spec_id": bad_spec})["error"]
+    with patch.object(Ctx, "load_payload", return_value={"issues": {"broken": []}}):
+        assert isinstance(et_mod._dispatch("list_broken_links", {}, ctx, conn), dict)
+
+    with patch("website_profiling.tools.export_custom.rows_from_tool_result", return_value=[]):
+        assert "/a" in _section_html_tool_result("H", {"pages": [{"path": "/a"}]})
+    _, err5 = validate_sections([{"type": "executive_summary"}, 42])
+    assert err5
+    import builtins
+    real_import = builtins.__import__
+
+    def _block_reportlab(name, globals=None, locals=None, fromlist=(), level=0):
+        if name == "reportlab" or (fromlist and "reportlab" in name):
+            raise ImportError("no reportlab")
+        return real_import(name, globals, locals, fromlist, level)
+
+    with patch("builtins.__import__", side_effect=_block_reportlab):
+        with pytest.raises(RuntimeError, match="reportlab"):
+            from website_profiling.tools.export_custom import render_custom_report_pdf as _pdf
+            _pdf("<html></html>", "T")
+
+    clusters_only_bad = ["bad", "bad2"]
+    with patch.object(Ctx, "load_payload", return_value={"content_duplicates": clusters_only_bad}):
+        assert content_mod.get_duplicate_cluster(conn, ctx, {"url": "https://ex.com/a"})["error"]
+
+    assert google_mod.get_ga4_page_metrics(conn, ctx, {"path": "/home"})["error"]
+    with patch.object(Ctx, "load_google", return_value={"ga4": {}}):
+        assert google_mod.get_ga4_page_metrics(conn, ctx, {"path": "/home"})["missing"]
+    with patch.object(Ctx, "load_google", return_value={"ga4": {"top_pages": [{"path": "home", "sessions": 1}]}}):
+        assert google_mod.get_ga4_page_metrics(conn, ctx, {"path": "home"})["metrics"]["sessions"] == 1
+
+    bad_hist_row = MagicMock()
+    bad_hist_row.__getitem__ = lambda self, i: [42, datetime.now(timezone.utc), 1, 80][i]
+    conn.execute = MagicMock(return_value=MagicMock(fetchall=MagicMock(return_value=[bad_hist_row])))
+    assert health_mod.get_category_health_history(conn, ctx, {})["count"] == 1
+
+    with patch.object(Ctx, "load_keywords", return_value=None):
+        assert kw_mod._keyword_list_tool(conn, ctx, {}, "semantic_keyword_clusters", "clusters")["error"]
+    with patch.object(Ctx, "load_keywords", return_value={"rows": []}), patch.object(
+        Ctx, "load_payload", return_value={"semantic_keyword_clusters": [{"keyword": "a"}]},
+    ):
+        assert kw_mod._keyword_list_tool(conn, ctx, {}, "semantic_keyword_clusters", "clusters")["total"] == 1
+    assert kw_mod.list_keywords_by_action(conn, Ctx(property_id=None), {"recommended_action": "improve"})["error"]
+    with patch.object(Ctx, "load_keywords", return_value={"rows": [{"keyword": "a", "gsc_position": 15}]}):
+        assert kw_mod.list_keywords_by_position(conn, ctx, {"min_position": 1, "max_position": 10})["total"] == 0
+
+    with patch.object(Ctx, "load_payload", return_value={}):
+        assert lh_mod.list_slow_pages(conn, ctx, {})["error"]
+        assert lh_mod.list_lighthouse_poor_seo_pages(conn, ctx, {})["error"]
+        assert lh_mod.list_lighthouse_cwv_failures(conn, ctx, {})["error"]
+    assert lh_mod._extract_lh_score({"scores": {"x": "bad"}, "category_scores": {"x": "bad"}, "median_metrics": {"x": "bad"}}, "x", "x_score") is None
+    assert lh_mod._extract_lh_score({"accessibility": "bad"}, "accessibility") is None
+
+    with patch.object(Ctx, "load_payload", return_value={"inlink_anchor_matrix": "bad"}):
+        assert links_mod.get_inlink_anchors(conn, ctx, {})["total"] == 0
+
+    snap_dict = MagicMock()
+    snap_dict.__getitem__ = lambda self, i: [80, datetime.now(timezone.utc), 1, {"High": 1}][i]
+    conn.execute = MagicMock(return_value=MagicMock(fetchone=MagicMock(return_value=snap_dict)))
+    with patch("website_profiling.tools.audit_tools.llm_tools.list_properties_public", return_value=[{"id": 1}]):
+        assert llm_mod.get_portfolio_summary(conn, ctx, {})["properties"][0]["issue_counts"] == {"High": 1}
+
+    with patch.object(Ctx, "load_payload", return_value={"categories": ["bad", {"id": "c", "issues": ["bad", {"message": "m", "llm_recommendation": "fix"}]}]}):
+        assert rex_mod.list_issues_with_ai_fixes(conn, ctx, {})["total"] == 1
+
+    with patch.object(Ctx, "load_payload", return_value={}):
+        assert report_mod.get_report_summary(conn, ctx, {})["error"]
+        assert report_mod.list_issues(conn, ctx, {"limit": "bad"})["total"] == 0
+    with patch.object(Ctx, "load_payload", return_value={"categories": [{"id": "c", "issues": [{"priority": "High", "message": "m"}]}]}):
+        assert report_mod.search_issues(conn, ctx, {"priority": "high"})["total"] == 1
+    with patch.object(Ctx, "load_payload", return_value={"categories": []}):
+        assert report_mod.list_issues(conn, ctx, {"limit": "bad"})["total"] == 0
+
+    with patch.object(Ctx, "load_payload", return_value={"categories": ["bad"]}):
+        assert issues_mod.get_category_issues(conn, ctx, {"category_id": "c"})["error"]
+
+    assert ops_mod.get_log_top_paths(conn, Ctx(property_id=None), {})["error"]
+    assert ops_mod.list_log_only_paths(conn, Ctx(property_id=None), {})["error"]
+    assert ops_mod.list_crawl_only_paths(conn, Ctx(property_id=None), {})["error"]
+    assert ops_mod.get_log_googlebot_stats(conn, Ctx(property_id=None), {})["error"]
+    assert ops_mod._parse_analysis_field({"k": 1}) == {"k": 1}
+    assert ops_mod._parse_analysis_field([1, 2]) == {}
+    with patch("website_profiling.tools.audit_tools.ops._load_log_analysis", return_value=None):
+        assert ops_mod.list_log_only_paths(conn, ctx, {})["missing"]
+        assert ops_mod.list_crawl_only_paths(conn, ctx, {})["missing"]
+        assert ops_mod.get_log_googlebot_stats(conn, ctx, {})["missing"]
+
+    diag_payload = {"lighthouse_diagnostics": [
+        "bad",
+        *[{"lighthouse_audit_id": "uses-optimized-images", "title": "T", "url": f"https://ex.com/{i}"} for i in range(10)],
+    ]}
+    assert len(img_mod._lighthouse_image_previews(diag_payload, limit=3)) == 3
+    dup_links = {"links": [
+        {"url": "https://ex.com", "og_image": "https://cdn.ex.com/same.png"},
+        {"url": "https://ex.com", "og_image": "https://cdn.ex.com/same.png"},
+    ]}
+    with patch.object(Ctx, "load_payload", return_value=dup_links):
+        assert img_mod.list_site_image_urls(conn, ctx, {})["total"] == 1
+    empty_inv = {"image_inventory": [], "image_inventory_summary": {"unoptimized_min_kb": 200}}
+    with patch.object(Ctx, "load_payload", return_value=empty_inv):
+        assert img_mod.list_unoptimized_images(conn, ctx, {})["inventory_available"] is False
+    mixed_inv = {"image_inventory": [
+        {"url": "https://cdn.ex.com/skip", "size_bytes": None, "content_type": "image/png"},
+        {"url": "https://cdn.ex.com/small", "size_bytes": 100, "content_type": "image/png"},
+    ], "image_inventory_summary": {"unoptimized_min_kb": 200}}
+    with patch.object(Ctx, "load_payload", return_value=mixed_inv):
+        assert img_mod.list_unoptimized_images(conn, ctx, {})["total"] == 0
+
+    rows, _ = wb_mod._custom_field_rows(["bad", {"url": "https://ex.com", "custom_fields": '{"a":"1"}'}])
+    assert len(rows) == 1
diff --git a/web/app/api/ai/fix-suggestion/route.ts b/web/app/api/ai/fix-suggestion/route.ts
new file mode 100644
index 00000000..054df7a7
--- /dev/null
+++ b/web/app/api/ai/fix-suggestion/route.ts
@@ -0,0 +1,75 @@
+import { NextResponse, type NextRequest } from 'next/server';
+import { spawn } from 'child_process';
+import { getRepoRoot, getPipelineSpawnEnv } from '@/server/pipelineSpawnEnv';
+import { resolvePythonExecutable, parsePythonJsonStdout } from '@/server/resolvePython';
+import type { ApiRouteHandler } from '@/types/api';
+
+export const runtime = 'nodejs';
+export const dynamic = 'force-dynamic';
+
+const PYTHON_SCRIPT = `
+import json, sys
+from website_profiling.llm.fix_suggestions import generate_fix_suggestion
+payload = json.load(sys.stdin)
+print(json.dumps(generate_fix_suggestion(payload, refresh=bool(payload.get("refresh")))))
+`;
+
+/**
+ * POST /api/ai/fix-suggestion — on-demand LLM fix for any audit surface.
+ */
+export const POST: ApiRouteHandler = async (request: NextRequest): Promise<Response> => {
+  let body: {
+    source?: string;
+    message?: string;
+    url?: string;
+    refresh?: boolean;
+    context?: Record<string, unknown>;
+    priority?: string;
+    category?: string;
+    recommendation?: string;
+    type?: string;
+  };
+  try {
+    body = await request.json();
+  } catch {
+    return NextResponse.json({ error: 'Invalid JSON' }, { status: 400 });
+  }
+  const message = String(body.message || '').trim();
+  if (!message) {
+    return NextResponse.json({ error: 'message required' }, { status: 400 });
+  }
+
+  const repoRoot = getRepoRoot();
+  const pythonExe = resolvePythonExecutable(null, repoRoot);
+  const payload = {
+    source: body.source || 'issue',
+    message,
+    url: body.url,
+    refresh: body.refresh,
+    context: body.context,
+    priority: body.priority,
+    category: body.category,
+    recommendation: body.recommendation,
+    type: body.type,
+  };
+
+  return new Promise<Response>((resolve) => {
+    const proc = spawn(pythonExe, ['-c', PYTHON_SCRIPT], {
+      cwd: repoRoot,
+      env: getPipelineSpawnEnv(repoRoot),
+      shell: false,
+    });
+    let stdout = '';
+    proc.stdout?.on('data', (c: Buffer | string) => { stdout += c.toString(); });
+    proc.stdin?.write(JSON.stringify(payload));
+    proc.stdin?.end();
+    proc.on('close', (code) => {
+      const parsed = parsePythonJsonStdout(stdout);
+      if (code === 0 && parsed) {
+        resolve(NextResponse.json(parsed));
+        return;
+      }
+      resolve(NextResponse.json({ error: stdout.trim() || 'Fix suggestion failed' }, { status: 500 }));
+    });
+  });
+};
diff --git a/web/app/api/filters/route.ts b/web/app/api/filters/route.ts
new file mode 100644
index 00000000..60dc53c5
--- /dev/null
+++ b/web/app/api/filters/route.ts
@@ -0,0 +1,37 @@
+import { NextResponse, type NextRequest } from 'next/server';
+import { deleteSavedFilter, listSavedFilters, upsertSavedFilter } from '@/server/savedFiltersDb';
+
+export const runtime = 'nodejs';
+export const dynamic = 'force-dynamic';
+
+export async function GET(request: NextRequest) {
+  const propertyId = Number(request.nextUrl.searchParams.get('propertyId') || 0);
+  if (!propertyId) {
+    return NextResponse.json({ error: 'propertyId required' }, { status: 400 });
+  }
+  const filters = await listSavedFilters(propertyId);
+  return NextResponse.json({ filters });
+}
+
+export async function POST(request: NextRequest) {
+  const body = await request.json().catch(() => ({}));
+  const propertyId = Number(body.propertyId || 0);
+  const name = String(body.name || '').trim();
+  const filterJson = (body.filterJson && typeof body.filterJson === 'object') ? body.filterJson : {};
+  if (!propertyId || !name) {
+    return NextResponse.json({ error: 'propertyId and name required' }, { status: 400 });
+  }
+  await upsertSavedFilter(propertyId, name, filterJson);
+  return NextResponse.json({ ok: true });
+}
+
+export async function DELETE(request: NextRequest) {
+  const body = await request.json().catch(() => ({}));
+  const propertyId = Number(body.propertyId || 0);
+  const name = String(body.name || '').trim();
+  if (!propertyId || !name) {
+    return NextResponse.json({ error: 'propertyId and name required' }, { status: 400 });
+  }
+  await deleteSavedFilter(propertyId, name);
+  return NextResponse.json({ ok: true });
+}
diff --git a/web/app/api/issues/fix-suggestion/route.ts b/web/app/api/issues/fix-suggestion/route.ts
index f261186c..95a19656 100644
--- a/web/app/api/issues/fix-suggestion/route.ts
+++ b/web/app/api/issues/fix-suggestion/route.ts
@@ -7,8 +7,15 @@ import type { ApiRouteHandler } from '@/types/api';
 export const runtime = 'nodejs';
 export const dynamic = 'force-dynamic';
 
+const PYTHON_SCRIPT = `
+import json, sys
+from website_profiling.llm.fix_suggestions import generate_fix_suggestion
+payload = json.load(sys.stdin)
+print(json.dumps(generate_fix_suggestion(payload, refresh=bool(payload.get("refresh")))))
+`;
+
 /**
- * POST /api/issues/fix-suggestion — on-demand LLM fix for one issue.
+ * POST /api/issues/fix-suggestion — legacy alias for issue-source fix suggestions.
  */
 export const POST: ApiRouteHandler = async (request: NextRequest): Promise<Response> => {
   let body: {
@@ -32,32 +39,26 @@ export const POST: ApiRouteHandler = async (request: NextRequest): Promise<Respo
 
   const repoRoot = getRepoRoot();
   const pythonExe = resolvePythonExecutable(null, repoRoot);
-  const script = `
-import json, sys
-from website_profiling.llm.issue_fixes import generate_issue_fix_suggestion
-payload = json.load(sys.stdin)
-print(json.dumps(generate_issue_fix_suggestion(payload, refresh=bool(payload.get("refresh")))))
-`;
+  const payload = {
+    source: 'issue',
+    message,
+    url: body.url,
+    priority: body.priority,
+    category: body.category,
+    recommendation: body.recommendation,
+    type: body.type,
+    refresh: body.refresh,
+  };
 
   return new Promise<Response>((resolve) => {
-    const proc = spawn(pythonExe, ['-c', script], {
+    const proc = spawn(pythonExe, ['-c', PYTHON_SCRIPT], {
       cwd: repoRoot,
       env: getPipelineSpawnEnv(repoRoot),
       shell: false,
     });
     let stdout = '';
     proc.stdout?.on('data', (c: Buffer | string) => { stdout += c.toString(); });
-    proc.stdin?.write(
-      JSON.stringify({
-        message,
-        url: body.url,
-        priority: body.priority,
-        category: body.category,
-        recommendation: body.recommendation,
-        type: body.type,
-        refresh: body.refresh,
-      }),
-    );
+    proc.stdin?.write(JSON.stringify(payload));
     proc.stdin?.end();
     proc.on('close', (code) => {
       const parsed = parsePythonJsonStdout(stdout);
diff --git a/web/app/api/keywords/competitor-import/route.ts b/web/app/api/keywords/competitor-import/route.ts
new file mode 100644
index 00000000..fd14ea12
--- /dev/null
+++ b/web/app/api/keywords/competitor-import/route.ts
@@ -0,0 +1,65 @@
+import { NextResponse, type NextRequest } from 'next/server';
+import { spawn } from 'child_process';
+import path from 'path';
+import { forbiddenIfNotLocal } from '@/server/localOnly';
+import { resolvePythonExecutable } from '@/server/resolvePython';
+import { withDb } from '@/server/db';
+
+export const runtime = 'nodejs';
+export const dynamic = 'force-dynamic';
+
+const REPO_ROOT = process.env.WEBSITE_PROFILING_ROOT || path.resolve(process.cwd(), '..');
+
+export async function POST(request: NextRequest) {
+  const denied = forbiddenIfNotLocal(request);
+  if (denied) return denied;
+
+  const body = await request.json().catch(() => ({}));
+  const propertyId = Number(body.propertyId || 0);
+  const competitor = String(body.competitor || '').trim();
+  const csvText = String(body.csvText || '');
+  if (!propertyId || !competitor || !csvText.trim()) {
+    return NextResponse.json({ error: 'propertyId, competitor, and csvText required' }, { status: 400 });
+  }
+
+  const python = resolvePythonExecutable(process.env.PYTHON, REPO_ROOT);
+  const script = `
+import json, sys
+from website_profiling.integrations.keywords.competitor_csv import parse_competitor_keyword_csv
+rows = parse_competitor_keyword_csv(sys.stdin.read(), competitor=sys.argv[1])
+print(json.dumps({"count": len(rows), "rows": rows[:500]}))
+`;
+  return new Promise<Response>((resolve) => {
+    const proc = spawn(python, ['-c', script, competitor], {
+      cwd: REPO_ROOT,
+      env: { ...process.env, PYTHONPATH: path.join(REPO_ROOT, 'src') },
+    });
+    let out = '';
+    let err = '';
+    proc.stdout.on('data', (c) => { out += c.toString(); });
+    proc.stderr.on('data', (c) => { err += c.toString(); });
+    proc.stdin.write(csvText);
+    proc.stdin.end();
+    proc.on('close', (code) => {
+      if (code !== 0) {
+        resolve(NextResponse.json({ error: err.trim() || 'import failed' }, { status: 500 }));
+        return;
+      }
+      try {
+        const parsed = JSON.parse(out) as { count?: number; rows?: unknown[] };
+        const rows = Array.isArray(parsed.rows) ? parsed.rows : [];
+        void withDb(async (client) => {
+          await client.query(
+            `INSERT INTO pipeline_config (key, value, is_unknown, updated_at)
+             VALUES ('competitor_keyword_gap_json', $1, false, now())
+             ON CONFLICT (key) DO UPDATE SET value = EXCLUDED.value, updated_at = now()`,
+            [JSON.stringify(rows)],
+          );
+        }).catch(() => {});
+        resolve(NextResponse.json(parsed));
+      } catch {
+        resolve(NextResponse.json({ error: 'invalid response' }, { status: 500 }));
+      }
+    });
+  });
+}
diff --git a/web/app/api/report/export-sitemap/route.ts b/web/app/api/report/export-sitemap/route.ts
new file mode 100644
index 00000000..2b0b130b
--- /dev/null
+++ b/web/app/api/report/export-sitemap/route.ts
@@ -0,0 +1,60 @@
+import { NextResponse, type NextRequest } from 'next/server';
+import { spawn } from 'child_process';
+import path from 'path';
+import { forbiddenIfNotLocal } from '@/server/localOnly';
+import { resolvePythonExecutable } from '@/server/resolvePython';
+import type { ApiRouteHandler } from '@/types/api';
+
+export const runtime = 'nodejs';
+export const dynamic = 'force-dynamic';
+
+const REPO_ROOT = process.env.WEBSITE_PROFILING_ROOT || path.resolve(process.cwd(), '..');
+
+const SITEMAP_SCRIPT = `
+import sys
+from website_profiling.db import db_session
+from website_profiling.db.report_store import read_report_payload
+from website_profiling.tools.export_sitemap import build_sitemap_xml
+
+rid = int(sys.argv[1]) if sys.argv[1] != 'latest' else None
+with db_session() as conn:
+    payload = read_report_payload(conn, report_id=rid)
+if not payload:
+    raise SystemExit('no report found')
+print(build_sitemap_xml(payload), end='')
+`;
+
+export const GET: ApiRouteHandler = async (request: NextRequest): Promise<Response> => {
+  const denied = forbiddenIfNotLocal(request);
+  if (denied) return denied;
+
+  const reportId = request.nextUrl.searchParams.get('reportId');
+  const python = resolvePythonExecutable(process.env.PYTHON, REPO_ROOT);
+  const ridArg = reportId && /^\d+$/.test(reportId) ? reportId : 'latest';
+
+  return new Promise<Response>((resolve) => {
+    const proc = spawn(python, ['-c', SITEMAP_SCRIPT, ridArg], {
+      cwd: REPO_ROOT,
+      env: { ...process.env, PYTHONPATH: path.join(REPO_ROOT, 'src'), PYTHONIOENCODING: 'utf-8' },
+    });
+    let out = '';
+    let err = '';
+    proc.stdout.on('data', (c) => { out += c.toString(); });
+    proc.stderr.on('data', (c) => { err += c.toString(); });
+    proc.on('close', (code) => {
+      if (code !== 0) {
+        resolve(NextResponse.json({ error: err.trim() || 'Sitemap export failed' }, { status: 500 }));
+        return;
+      }
+      resolve(
+        new NextResponse(out, {
+          status: 200,
+          headers: {
+            'Content-Type': 'application/xml',
+            'Content-Disposition': 'attachment; filename="sitemap.xml"',
+          },
+        }),
+      );
+    });
+  });
+};
diff --git a/web/app/api/report/export-workbook/route.ts b/web/app/api/report/export-workbook/route.ts
new file mode 100644
index 00000000..fc740702
--- /dev/null
+++ b/web/app/api/report/export-workbook/route.ts
@@ -0,0 +1,69 @@
+import { NextResponse, type NextRequest } from 'next/server';
+import { spawn } from 'child_process';
+import path from 'path';
+import { forbiddenIfNotLocal } from '@/server/localOnly';
+import { resolvePythonExecutable } from '@/server/resolvePython';
+import type { ApiRouteHandler } from '@/types/api';
+
+export const runtime = 'nodejs';
+export const dynamic = 'force-dynamic';
+
+const REPO_ROOT = process.env.WEBSITE_PROFILING_ROOT || path.resolve(process.cwd(), '..');
+
+const WORKBOOK_SCRIPT = `
+import sys
+from website_profiling.db import db_session
+from website_profiling.db.report_store import read_report_payload
+from website_profiling.tools.export_crawl_workbook import build_crawl_workbook_zip
+
+rid = int(sys.argv[1]) if sys.argv[1] != 'latest' else None
+with db_session() as conn:
+    payload = read_report_payload(conn, report_id=rid)
+if not payload:
+    raise SystemExit('no report found')
+sys.stdout.buffer.write(build_crawl_workbook_zip(payload))
+`;
+
+export const GET: ApiRouteHandler = async (request: NextRequest): Promise<Response> => {
+  const denied = forbiddenIfNotLocal(request);
+  if (denied) return denied;
+
+  const reportId = request.nextUrl.searchParams.get('reportId');
+  const python = resolvePythonExecutable(process.env.PYTHON, REPO_ROOT);
+  const ridArg = reportId && /^\d+$/.test(reportId) ? reportId : 'latest';
+
+  return new Promise((resolve) => {
+    const proc = spawn(python, ['-c', WORKBOOK_SCRIPT, ridArg], {
+      cwd: REPO_ROOT,
+      env: {
+        ...process.env,
+        PYTHONPATH: path.join(REPO_ROOT, 'src'),
+        PYTHONIOENCODING: 'utf-8',
+      },
+    });
+    const chunks: Buffer[] = [];
+    let err = '';
+    proc.stdout.on('data', (c: Buffer | string) => {
+      chunks.push(Buffer.isBuffer(c) ? c : Buffer.from(c));
+    });
+    proc.stderr.on('data', (c) => {
+      err += c.toString();
+    });
+    proc.on('close', (code) => {
+      if (code !== 0) {
+        resolve(NextResponse.json({ error: err.trim() || 'Workbook export failed' }, { status: 500 }));
+        return;
+      }
+      const body = Buffer.concat(chunks);
+      resolve(
+        new NextResponse(body, {
+          status: 200,
+          headers: {
+            'Content-Type': 'application/zip',
+            'Content-Disposition': 'attachment; filename="crawl-workbook.zip"',
+          },
+        }),
+      );
+    });
+  });
+};
diff --git a/web/app/contacts/page.tsx b/web/app/contacts/page.tsx
new file mode 100644
index 00000000..a87bf3e5
--- /dev/null
+++ b/web/app/contacts/page.tsx
@@ -0,0 +1,7 @@
+'use client';
+
+import ReportShell from '@/ReportShell';
+
+export default function ContactsPage() {
+  return <ReportShell slug="contacts" />;
+}
diff --git a/web/app/subdomains/page.tsx b/web/app/subdomains/page.tsx
new file mode 100644
index 00000000..600fc97d
--- /dev/null
+++ b/web/app/subdomains/page.tsx
@@ -0,0 +1,7 @@
+'use client';
+
+import ReportShell from '@/ReportShell';
+
+export default function SubdomainsPage() {
+  return <ReportShell slug="subdomains" />;
+}
diff --git a/web/src/ReportShell.tsx b/web/src/ReportShell.tsx
index 6d592fce..01f567c9 100644
--- a/web/src/ReportShell.tsx
+++ b/web/src/ReportShell.tsx
@@ -26,6 +26,8 @@ import {
   FileDown,
   FileSearch,
   Terminal,
+  Globe2,
+  Contact2,
 } from 'lucide-react';
 import { UrlInspectorProvider } from './context/UrlInspectorContext';
 import AppShell from './components/AppShell';
@@ -72,6 +74,8 @@ const Traffic = dynamic(() => import('./views/Traffic'), { loading: () => viewLo
 const KeywordsExplorer = dynamic(() => import('./views/KeywordsExplorer'), { loading: () => viewLoading() });
 const ExportReport = dynamic(() => import('./views/ExportReport'), { loading: () => viewLoading() });
 const LogAnalyzer = dynamic(() => import('./views/LogAnalyzer'), { loading: () => viewLoading() });
+const Subdomains = dynamic(() => import('./views/Subdomains'), { loading: () => viewLoading() });
+const Contacts = dynamic(() => import('./views/Contacts'), { loading: () => viewLoading() });
 
 interface ReportShellReportContext {
   data: ReportPayload | null;
@@ -122,6 +126,8 @@ const VIEW_CONFIG: ViewConfigEntry[] = [
   { id: 'gallery', component: Gallery as ComponentType<CurrentViewProps>, icon: Images },
   { id: 'search-performance', component: SearchPerformance as ComponentType<CurrentViewProps>, icon: TrendingUp },
   { id: 'indexation', component: Indexation as ComponentType<CurrentViewProps>, icon: FileSearch },
+  { id: 'subdomains', component: Subdomains as ComponentType<CurrentViewProps>, icon: Globe2 },
+  { id: 'contacts', component: Contacts as ComponentType<CurrentViewProps>, icon: Contact2 },
   { id: 'backlinks', component: Backlinks as ComponentType<CurrentViewProps>, icon: Link2 },
   { id: 'traffic', component: Traffic as ComponentType<CurrentViewProps>, icon: BarChart2 },
   { id: 'keywords-explorer', component: KeywordsExplorer as ComponentType<CurrentViewProps>, icon: Key },
diff --git a/web/src/components/ai/AiSuggestionButton.tsx b/web/src/components/ai/AiSuggestionButton.tsx
new file mode 100644
index 00000000..c7419196
--- /dev/null
+++ b/web/src/components/ai/AiSuggestionButton.tsx
@@ -0,0 +1,82 @@
+'use client';
+
+import { useState, useCallback } from 'react';
+import { Loader2, Sparkles } from 'lucide-react';
+import { apiUrl } from '@/lib/publicBase';
+import { strings } from '@/lib/strings';
+import { useReadOnlySession } from '@/hooks/useReadOnlySession';
+import CopyBtn from '@/components/links/CopyBtn';
+import type { FixSuggestionRequest } from '@/types/fixSuggestion';
+import { extractFixText, type FixSuggestionResponse } from '@/types/fixSuggestion';
+
+export interface AiSuggestionButtonProps {
+  request: FixSuggestionRequest;
+  initialText?: string | null;
+  className?: string;
+}
+
+export default function AiSuggestionButton({ request, initialText = null, className = '' }: AiSuggestionButtonProps) {
+  const s = strings.components.aiSuggestion;
+  const { readOnly } = useReadOnlySession();
+  const [loading, setLoading] = useState(false);
+  const [text, setText] = useState<string | null>(
+    typeof initialText === 'string' && initialText.trim() ? initialText.trim() : null,
+  );
+  const [error, setError] = useState<string | null>(null);
+
+  const handleClick = useCallback(async () => {
+    if (readOnly) return;
+    setLoading(true);
+    setError(null);
+    try {
+      const res = await fetch(apiUrl('/ai/fix-suggestion'), {
+        method: 'POST',
+        headers: { 'Content-Type': 'application/json' },
+        body: JSON.stringify({
+          ...request,
+          refresh: !!text,
+        }),
+      });
+      const payload = (await res.json()) as FixSuggestionResponse & { error?: string };
+      if (!res.ok || payload.ok === false) {
+        throw new Error(payload.error || s.failed);
+      }
+      const fixText = extractFixText(payload) || s.empty;
+      setText(fixText);
+    } catch (e) {
+      setError(e instanceof Error ? e.message : s.failed);
+    } finally {
+      setLoading(false);
+    }
+  }, [readOnly, request, text, s.failed, s.empty]);
+
+  return (
+    <div className={`space-y-2 ${className}`.trim()}>
+      {!readOnly ? (
+        <button
+          type="button"
+          onClick={() => void handleClick()}
+          disabled={loading}
+          className="inline-flex items-center gap-1.5 rounded-md border border-fuchsia-500/30 bg-fuchsia-500/10 px-2.5 py-1 text-[10px] font-semibold text-fuchsia-800 dark:text-fuchsia-300 hover:bg-fuchsia-500/20 transition-colors disabled:opacity-60"
+        >
+          {loading ? (
+            <Loader2 className="h-3 w-3 animate-spin" aria-hidden />
+          ) : (
+            <Sparkles className="h-3 w-3" aria-hidden />
+          )}
+          {loading ? s.loading : text ? s.regenerate : s.button}
+        </button>
+      ) : null}
+      {error ? <p className="text-xs text-red-700 dark:text-red-400">{error}</p> : null}
+      {text ? (
+        <div className="flex items-start gap-2">
+          <p className="text-xs text-muted-foreground leading-relaxed flex-1 min-w-0">
+            <span className="text-fuchsia-700 dark:text-fuchsia-300 font-semibold">{s.label}: </span>
+            {text}
+          </p>
+          <CopyBtn text={text} className="shrink-0 mt-0.5" />
+        </div>
+      ) : null}
+    </div>
+  );
+}
diff --git a/web/src/components/compare/CompareTabPanels.tsx b/web/src/components/compare/CompareTabPanels.tsx
index d95d92a6..f1b1ce09 100644
--- a/web/src/components/compare/CompareTabPanels.tsx
+++ b/web/src/components/compare/CompareTabPanels.tsx
@@ -16,6 +16,8 @@ import { ScoreDelta } from '@/components/charts/ScoreDelta';
 import type { CompareMetricRow, ReportCompareSummary } from '@/lib/reportCompare';
 import { Card, Table, TableHead, TableHeadCell, TableBody, TableRow, TableCell, Badge } from '@/components';
 import { CompareMetricCard } from './CompareDeltaBadge';
+import AiSuggestionButton from '@/components/ai/AiSuggestionButton';
+import { buildCompareIssueContext, buildCompareSecurityContext } from '@/lib/fixSuggestionContext';
 
 function matchesQuery(text: string, q: string): boolean {
   return text.toLowerCase().includes(q);
@@ -133,10 +135,20 @@ export function CompareIssuesPanel({ compare, searchQuery, vc, emptyLabel }: Pan
                   <TableCell>
                     <KindBadge kind={row.kind} vc={vc} />
                   </TableCell>
-                  <TableCell className="text-sm">
+                  <TableCell className="text-sm align-top">
                     <span className="font-medium">{row.findingType}</span>
                     <span className="text-muted-foreground"> · {row.severity}</span>
                     <p className="text-xs text-muted-foreground mt-0.5">{row.message}</p>
+                    <AiSuggestionButton
+                      request={buildCompareSecurityContext({
+                        message: row.message,
+                        url: row.url,
+                        findingType: row.findingType,
+                        severity: row.severity,
+                        kind: row.kind,
+                      })}
+                      className="mt-2"
+                    />
                   </TableCell>
                   <TableCell className="font-mono text-xs break-all">{row.url}</TableCell>
                 </TableRow>
@@ -207,8 +219,13 @@ function IssueTable({
                 <Badge variant={row.priority === 'Critical' || row.priority === 'High' ? 'high' : 'medium'} label={row.priority} />
               </TableCell>
               <TableCell className="text-sm">{row.category}</TableCell>
-              <TableCell className="text-sm text-muted-foreground max-w-md">{row.message}</TableCell>
-              <TableCell className="font-mono text-xs break-all">{row.url}</TableCell>
+              <TableCell className="text-sm text-muted-foreground max-w-md align-top">
+                <div className="space-y-2">
+                  <span>{row.message}</span>
+                  <AiSuggestionButton request={buildCompareIssueContext(row)} />
+                </div>
+              </TableCell>
+              <TableCell className="font-mono text-xs break-all align-top">{row.url}</TableCell>
             </TableRow>
           ))}
         </TableBody>
diff --git a/web/src/components/compare/CompareUrlMetadataTable.tsx b/web/src/components/compare/CompareUrlMetadataTable.tsx
new file mode 100644
index 00000000..0c7e1cad
--- /dev/null
+++ b/web/src/components/compare/CompareUrlMetadataTable.tsx
@@ -0,0 +1,39 @@
+'use client';
+
+import { Table, TableBody, TableCell, TableHead, TableHeadCell, TableRow } from '@/components';
+import type { UrlMetadataChangeRow } from '@/lib/reportCompare';
+
+interface CompareUrlMetadataTableProps {
+  rows: UrlMetadataChangeRow[];
+  emptyLabel: string;
+}
+
+export default function CompareUrlMetadataTable({ rows, emptyLabel }: CompareUrlMetadataTableProps) {
+  if (!rows.length) {
+    return <p className="text-sm text-muted-foreground py-4">{emptyLabel}</p>;
+  }
+  return (
+    <div className="max-h-[min(480px,55vh)] overflow-y-auto border border-default rounded-lg">
+      <Table>
+        <TableHead sticky>
+          <TableRow>
+            <TableHeadCell>URL</TableHeadCell>
+            <TableHeadCell>Field</TableHeadCell>
+            <TableHeadCell>Baseline</TableHeadCell>
+            <TableHeadCell>Current</TableHeadCell>
+          </TableRow>
+        </TableHead>
+        <TableBody striped>
+          {rows.map((row) => (
+            <TableRow key={`${row.url}-${row.field}`}>
+              <TableCell className="font-mono text-xs break-all max-w-[220px]">{row.url}</TableCell>
+              <TableCell className="text-xs">{row.field}</TableCell>
+              <TableCell className="text-xs break-all max-w-[180px]">{row.baseline || '—'}</TableCell>
+              <TableCell className="text-xs break-all max-w-[180px]">{row.current || '—'}</TableCell>
+            </TableRow>
+          ))}
+        </TableBody>
+      </Table>
+    </div>
+  );
+}
diff --git a/web/src/components/content/RichResultsValidationPanel.tsx b/web/src/components/content/RichResultsValidationPanel.tsx
new file mode 100644
index 00000000..5b464ea6
--- /dev/null
+++ b/web/src/components/content/RichResultsValidationPanel.tsx
@@ -0,0 +1,94 @@
+'use client';
+
+import { useMemo } from 'react';
+import { Sparkles } from 'lucide-react';
+import { Card } from '@/components';
+import Badge from '@/components/Badge';
+import SortablePaginatedTable from '@/components/google/SortablePaginatedTable';
+import { strings, format } from '@/lib/strings';
+import type { RichResultsMeta, RichResultsValidationRow } from '@/types/report';
+import type { TableColumn } from '@/types/components';
+import AiSuggestionButton from '@/components/ai/AiSuggestionButton';
+import { buildRichResultsContext } from '@/lib/fixSuggestionContext';
+
+interface RichResultsValidationPanelProps {
+  rows: RichResultsValidationRow[];
+  meta?: RichResultsMeta | null;
+}
+
+function statusBadgeVariant(status: string): string {
+  switch (status) {
+    case 'pass':
+      return 'success';
+    case 'warning':
+      return 'medium';
+    case 'fail':
+    case 'error':
+      return 'critical';
+    default:
+      return 'info';
+  }
+}
+
+export default function RichResultsValidationPanel({ rows, meta }: RichResultsValidationPanelProps) {
+  const vca = strings.views.contentAnalytics;
+  const columns = useMemo((): TableColumn[] => [
+    { key: 'url', label: vca.richResultsColUrl },
+    {
+      key: 'status',
+      label: vca.richResultsColStatus,
+      render: (v) => (
+        <Badge variant={statusBadgeVariant(String(v || ''))} value={String(v || 'info')} />
+      ),
+    },
+    { key: 'provenance', label: vca.richResultsColSource },
+    { key: 'message', label: vca.richResultsColMessage },
+    {
+      key: '_ai',
+      label: '',
+      render: (_v, row) => (
+        <AiSuggestionButton request={buildRichResultsContext(row as unknown as RichResultsValidationRow)} />
+      ),
+    },
+  ], [vca]);
+
+  if (!rows.length) return null;
+
+  const heuristicOnly =
+    meta != null &&
+    (meta.heuristic_count ?? 0) > 0 &&
+    (meta.gsc_count ?? 0) === 0 &&
+    (meta.api_count ?? 0) === 0;
+
+  return (
+    <Card padding="tight" shadow>
+      <div className="flex items-center gap-2 mb-2">
+        <Sparkles className="h-4 w-4 text-violet-700 dark:text-violet-400" />
+        <h3 className="text-sm font-bold text-foreground">{vca.richResultsTitle}</h3>
+      </div>
+      <p className="text-xs text-muted-foreground mb-2">{vca.richResultsHint}</p>
+      {meta && (meta.checked ?? 0) > 0 ? (
+        <p className="text-xs text-muted-foreground mb-2">
+          {format(vca.richResultsMeta, {
+            gsc: meta.gsc_count ?? 0,
+            api: meta.api_count ?? 0,
+            heuristic: meta.heuristic_count ?? 0,
+          })}
+        </p>
+      ) : null}
+      {heuristicOnly ? (
+        <p className="text-xs text-amber-700 dark:text-amber-300 mb-4">{vca.richResultsUpgradeHint}</p>
+      ) : (
+        <div className="mb-4" />
+      )}
+      <SortablePaginatedTable
+        columns={columns}
+        rows={rows as unknown as Array<Record<string, unknown>>}
+        defaultSort="status"
+        rowKeyField="url"
+        emptyMessage={vca.richResultsEmpty}
+        paginationLabels={strings.views.keywordsExplorer.table}
+      />
+    </Card>
+  );
+}
diff --git a/web/src/components/issues/IssueAiFixButton.tsx b/web/src/components/issues/IssueAiFixButton.tsx
index 9061f72e..ab424a8b 100644
--- a/web/src/components/issues/IssueAiFixButton.tsx
+++ b/web/src/components/issues/IssueAiFixButton.tsx
@@ -1,11 +1,8 @@
 'use client';
 
-import { useState, useCallback } from 'react';
-import { Loader2, Sparkles } from 'lucide-react';
-import { apiUrl } from '@/lib/publicBase';
-import { strings } from '@/lib/strings';
+import AiSuggestionButton from '@/components/ai/AiSuggestionButton';
+import { buildIssueContext } from '@/lib/fixSuggestionContext';
 import type { ReportIssue } from '@/types';
-import { useReadOnlySession } from '@/hooks/useReadOnlySession';
 
 export interface IssueAiFixButtonProps {
   issue: ReportIssue;
@@ -13,67 +10,12 @@ export interface IssueAiFixButtonProps {
 }
 
 export default function IssueAiFixButton({ issue, category }: IssueAiFixButtonProps) {
-  const s = strings.views.issues.aiFix;
-  const { readOnly } = useReadOnlySession();
-  const [loading, setLoading] = useState(false);
-  const [text, setText] = useState<string | null>(
-    typeof issue.llm_recommendation === 'string' ? issue.llm_recommendation : null,
-  );
-  const [error, setError] = useState<string | null>(null);
-
-  const handleClick = useCallback(async () => {
-    if (readOnly) return;
-    setLoading(true);
-    setError(null);
-    try {
-      const res = await fetch(apiUrl('/issues/fix-suggestion'), {
-        method: 'POST',
-        headers: { 'Content-Type': 'application/json' },
-        body: JSON.stringify({
-          message: issue.message,
-          url: issue.url,
-          priority: issue.priority,
-          category,
-          recommendation: issue.recommendation,
-          type: issue.type || issue.finding_type,
-          refresh: !!text,
-        }),
-      });
-      const payload = await res.json();
-      if (!res.ok) throw new Error(payload.error || s.failed);
-      const fix = payload.fix as { fix?: string } | undefined;
-      setText(String(fix?.fix || payload.fix || '').trim() || s.empty);
-    } catch (e) {
-      setError(e instanceof Error ? e.message : s.failed);
-    } finally {
-      setLoading(false);
-    }
-  }, [issue, category, text, readOnly, s.failed, s.empty]);
-
+  const initialText =
+    typeof issue.llm_recommendation === 'string' ? issue.llm_recommendation : null;
   return (
-    <div className="space-y-2">
-      {!readOnly ? (
-      <button
-        type="button"
-        onClick={() => void handleClick()}
-        disabled={loading}
-        className="inline-flex items-center gap-1.5 rounded-md border border-fuchsia-500/30 bg-fuchsia-500/10 px-2.5 py-1 text-[10px] font-semibold text-fuchsia-800 dark:text-fuchsia-300 hover:bg-fuchsia-500/20 transition-colors disabled:opacity-60"
-      >
-        {loading ? (
-          <Loader2 className="h-3 w-3 animate-spin" aria-hidden />
-        ) : (
-          <Sparkles className="h-3 w-3" aria-hidden />
-        )}
-        {loading ? s.loading : text ? s.regenerate : s.button}
-      </button>
-      ) : null}
-      {error ? <p className="text-xs text-red-700 dark:text-red-400">{error}</p> : null}
-      {text ? (
-        <p className="text-xs text-muted-foreground leading-relaxed">
-          <span className="text-fuchsia-700 dark:text-fuchsia-300 font-semibold">{s.label}: </span>
-          {text}
-        </p>
-      ) : null}
-    </div>
+    <AiSuggestionButton
+      request={buildIssueContext(issue, category)}
+      initialText={initialText}
+    />
   );
 }
diff --git a/web/src/components/keywordsExplorer/CompetitorKeywordGapPanel.tsx b/web/src/components/keywordsExplorer/CompetitorKeywordGapPanel.tsx
new file mode 100644
index 00000000..35c57fab
--- /dev/null
+++ b/web/src/components/keywordsExplorer/CompetitorKeywordGapPanel.tsx
@@ -0,0 +1,50 @@
+'use client';
+
+import { useMemo } from 'react';
+import SortablePaginatedTable from '@/components/google/SortablePaginatedTable';
+import { strings } from '@/lib/strings';
+import type { CompetitorKeywordGapRow } from '@/types/report';
+import type { TableColumn } from '@/types/components';
+import AiSuggestionButton from '@/components/ai/AiSuggestionButton';
+import { buildKeywordGapContext } from '@/lib/fixSuggestionContext';
+
+interface CompetitorKeywordGapPanelProps {
+  rows: CompetitorKeywordGapRow[];
+}
+
+export default function CompetitorKeywordGapPanel({ rows }: CompetitorKeywordGapPanelProps) {
+  const ke = strings.views.keywordsExplorer.competitorGap;
+  const columns = useMemo((): TableColumn[] => [
+    { key: 'keyword', label: ke.colKeyword },
+    { key: 'competitor', label: ke.colCompetitor },
+    { key: 'volume', label: ke.colVolume },
+    { key: 'position', label: ke.colPosition },
+    { key: 'url', label: ke.colUrl },
+    {
+      key: '_ai',
+      label: '',
+      render: (_v, row) => (
+        <AiSuggestionButton request={buildKeywordGapContext(row as unknown as CompetitorKeywordGapRow)} />
+      ),
+    },
+  ], [ke]);
+
+  if (!rows.length) {
+    return <p className="text-sm text-muted-foreground py-4">{ke.empty}</p>;
+  }
+
+  return (
+    <div className="p-4 border-t border-default">
+      <h3 className="text-sm font-semibold text-foreground mb-1">{ke.title}</h3>
+      <p className="text-xs text-muted-foreground mb-4">{ke.hint}</p>
+      <SortablePaginatedTable
+        columns={columns}
+        rows={rows as unknown as Array<Record<string, unknown>>}
+        defaultSort="volume"
+        rowKeyField="keyword"
+        emptyMessage={ke.empty}
+        paginationLabels={strings.views.keywordsExplorer.table}
+      />
+    </div>
+  );
+}
diff --git a/web/src/components/keywordsExplorer/CompetitorKeywordImport.tsx b/web/src/components/keywordsExplorer/CompetitorKeywordImport.tsx
new file mode 100644
index 00000000..1733ab5a
--- /dev/null
+++ b/web/src/components/keywordsExplorer/CompetitorKeywordImport.tsx
@@ -0,0 +1,68 @@
+'use client';
+
+import { useState } from 'react';
+import { Upload } from 'lucide-react';
+import { Button, Card } from '@/components';
+import { apiUrl } from '@/lib/publicBase';
+
+interface CompetitorKeywordImportProps {
+  propertyId: number;
+  onImported?: (count: number) => void;
+}
+
+export default function CompetitorKeywordImport({ propertyId, onImported }: CompetitorKeywordImportProps) {
+  const [competitor, setCompetitor] = useState('');
+  const [csvText, setCsvText] = useState('');
+  const [status, setStatus] = useState('');
+  const [busy, setBusy] = useState(false);
+
+  const handleImport = async () => {
+    if (!propertyId || !competitor.trim() || !csvText.trim()) return;
+    setBusy(true);
+    setStatus('');
+    try {
+      const res = await fetch(apiUrl('/keywords/competitor-import'), {
+        method: 'POST',
+        headers: { 'Content-Type': 'application/json' },
+        body: JSON.stringify({ propertyId, competitor: competitor.trim(), csvText }),
+      });
+      const body = await res.json();
+      if (!res.ok) throw new Error(body.error || 'Import failed');
+      setStatus(`Imported ${body.count ?? 0} keywords`);
+      onImported?.(body.count ?? 0);
+    } catch (e) {
+      setStatus(e instanceof Error ? e.message : 'Import failed');
+    } finally {
+      setBusy(false);
+    }
+  };
+
+  return (
+    <Card className="p-4 space-y-3">
+      <div className="flex items-center gap-2 text-sm font-semibold">
+        <Upload className="h-4 w-4" />
+        Competitor keyword CSV (Ahrefs / Semrush export)
+      </div>
+      <input
+        type="text"
+        placeholder="Competitor domain"
+        value={competitor}
+        onChange={(e) => setCompetitor(e.target.value)}
+        className="w-full rounded-lg border border-default bg-background px-3 py-2 text-sm"
+      />
+      <textarea
+        placeholder="Paste CSV export…"
+        value={csvText}
+        onChange={(e) => setCsvText(e.target.value)}
+        rows={6}
+        className="w-full rounded-lg border border-default bg-background px-3 py-2 text-sm font-mono"
+      />
+      <div className="flex items-center gap-3">
+        <Button type="button" onClick={() => void handleImport()} disabled={busy}>
+          Import
+        </Button>
+        {status ? <span className="text-xs text-muted-foreground">{status}</span> : null}
+      </div>
+    </Card>
+  );
+}
diff --git a/web/src/components/keywordsExplorer/ContentTemplatesPanel.tsx b/web/src/components/keywordsExplorer/ContentTemplatesPanel.tsx
new file mode 100644
index 00000000..aff99aa1
--- /dev/null
+++ b/web/src/components/keywordsExplorer/ContentTemplatesPanel.tsx
@@ -0,0 +1,166 @@
+'use client';
+
+import { useCallback, useState } from 'react';
+import { FileText, Loader2, X } from 'lucide-react';
+import { apiUrl } from '@/lib/publicBase';
+import { strings } from '@/lib/strings';
+import type { KeywordRow } from '@/types/components';
+import { useReadOnlySession } from '@/hooks/useReadOnlySession';
+import { Card, Button } from '@/components';
+
+const TEMPLATES = [
+  {
+    id: 'blog',
+    title: 'Blog post',
+    outline: ['H1 with primary keyword', 'Intro (150 words)', 'H2 sections (3–5)', 'FAQ block', 'Internal links (3+)'],
+  },
+  {
+    id: 'landing',
+    title: 'Landing page',
+    outline: ['Hero + value prop', 'Social proof', 'Feature bullets', 'Primary CTA', 'Schema: Organization or Product'],
+  },
+  {
+    id: 'comparison',
+    title: 'Comparison page',
+    outline: ['H1: X vs Y', 'Summary table', 'Pros/cons per option', 'Recommendation', 'FAQ schema'],
+  },
+] as const;
+
+interface ContentBriefResult {
+  keyword?: string;
+  summary?: string;
+  provenance?: string;
+}
+
+interface ContentTemplatesPanelProps {
+  defaultKeyword?: string;
+  clusterRows?: KeywordRow[];
+}
+
+export default function ContentTemplatesPanel({ defaultKeyword = '', clusterRows = [] }: ContentTemplatesPanelProps) {
+  const s = strings.views.keywordsExplorer.contentBrief;
+  const { readOnly } = useReadOnlySession();
+  const [activeTemplate, setActiveTemplate] = useState<(typeof TEMPLATES)[number] | null>(null);
+  const [keyword, setKeyword] = useState(defaultKeyword);
+  const [loading, setLoading] = useState(false);
+  const [brief, setBrief] = useState<ContentBriefResult | null>(null);
+  const [error, setError] = useState<string | null>(null);
+
+  const openTemplate = useCallback(
+    (template: (typeof TEMPLATES)[number]) => {
+      if (readOnly) return;
+      setActiveTemplate(template);
+      setBrief(null);
+      setError(null);
+      if (!keyword.trim() && defaultKeyword) setKeyword(defaultKeyword);
+    },
+    [readOnly, keyword, defaultKeyword],
+  );
+
+  const generateBrief = useCallback(async () => {
+    if (readOnly || !activeTemplate) return;
+    const kw = keyword.trim();
+    if (!kw) {
+      setError('Enter a target keyword first.');
+      return;
+    }
+    setLoading(true);
+    setError(null);
+    setBrief(null);
+    try {
+      const res = await fetch(apiUrl('/keywords/content-brief'), {
+        method: 'POST',
+        headers: { 'Content-Type': 'application/json' },
+        body: JSON.stringify({
+          keyword: `${kw} (${activeTemplate.title} template)`,
+          rows: clusterRows.slice(0, 20),
+          templateId: activeTemplate.id,
+        }),
+      });
+      const payload = await res.json();
+      if (!res.ok) throw new Error(payload.error || s.failed);
+      setBrief((payload.brief || null) as ContentBriefResult | null);
+    } catch (e) {
+      setError(e instanceof Error ? e.message : s.failed);
+    } finally {
+      setLoading(false);
+    }
+  }, [readOnly, activeTemplate, keyword, clusterRows, s.failed]);
+
+  return (
+    <>
+      <div className="grid gap-4 sm:grid-cols-3">
+        {TEMPLATES.map((t) => (
+          <Card key={t.id} className="p-4 flex flex-col">
+            <h3 className="text-sm font-semibold">{t.title}</h3>
+            <ul className="mt-2 text-xs text-muted-foreground space-y-1 flex-1 list-disc pl-4">
+              {t.outline.map((line) => (
+                <li key={line}>{line}</li>
+              ))}
+            </ul>
+            <Button
+              type="button"
+              variant="secondary"
+              className="mt-3 !py-1.5 !text-xs"
+              onClick={() => openTemplate(t)}
+              disabled={readOnly}
+            >
+              Use template
+            </Button>
+          </Card>
+        ))}
+      </div>
+      {activeTemplate ? (
+        <div
+          className="fixed inset-0 z-[80] flex items-center justify-center bg-black/50 p-4"
+          role="dialog"
+          aria-modal="true"
+          aria-labelledby="content-template-title"
+        >
+          <div className="w-full max-w-lg rounded-xl border border-default bg-brand-800 shadow-xl">
+            <div className="flex items-center justify-between border-b border-default px-4 py-3">
+              <h3 id="content-template-title" className="text-sm font-semibold text-bright">
+                {activeTemplate.title}
+              </h3>
+              <button
+                type="button"
+                onClick={() => setActiveTemplate(null)}
+                className="rounded-md p-1 text-muted-foreground hover:text-foreground"
+                aria-label={s.close}
+              >
+                <X className="h-4 w-4" aria-hidden />
+              </button>
+            </div>
+            <div className="px-4 py-4 space-y-3 text-sm">
+              <label className="block text-xs text-muted-foreground">
+                Target keyword
+                <input
+                  type="text"
+                  value={keyword}
+                  onChange={(e) => setKeyword(e.target.value)}
+                  className="mt-1 w-full rounded-md border border-default bg-brand-900 px-3 py-2 text-sm text-foreground"
+                  placeholder="primary keyword"
+                />
+              </label>
+              <Button type="button" variant="primary" className="!text-xs" onClick={() => void generateBrief()} disabled={loading || readOnly}>
+                {loading ? <Loader2 className="h-3.5 w-3.5 animate-spin" aria-hidden /> : <FileText className="h-3.5 w-3.5" aria-hidden />}
+                {loading ? s.loading : s.buttonLabel}
+              </Button>
+              {error ? <p className="text-red-700 dark:text-red-400 text-xs">{error}</p> : null}
+              {brief?.summary ? (
+                <>
+                  <pre className="whitespace-pre-wrap text-xs text-muted-foreground leading-relaxed font-sans">
+                    {brief.summary}
+                  </pre>
+                  {brief.provenance ? (
+                    <p className="text-[10px] text-muted-foreground">{s.provenance}: {brief.provenance}</p>
+                  ) : null}
+                </>
+              ) : null}
+            </div>
+          </div>
+        </div>
+      ) : null}
+    </>
+  );
+}
diff --git a/web/src/components/keywordsExplorer/KeywordPanels.tsx b/web/src/components/keywordsExplorer/KeywordPanels.tsx
index 04049bc4..2b08a2f3 100644
--- a/web/src/components/keywordsExplorer/KeywordPanels.tsx
+++ b/web/src/components/keywordsExplorer/KeywordPanels.tsx
@@ -10,6 +10,8 @@ import { useSearchParams } from 'next/navigation';
 import { apiUrl } from '../../lib/publicBase';
 import { buildLinksInspectHref } from '../../lib/reportNav';
 import UrlInspectorButton from '@/components/UrlInspectorButton';
+import AiSuggestionButton from '@/components/ai/AiSuggestionButton';
+import { buildCannibalisationContext, buildMisalignmentContext } from '@/lib/fixSuggestionContext';
 import { strings, format } from '../../lib/strings';
 import { Card } from '../index';
 import CopyBtn from '../links/CopyBtn';
@@ -126,6 +128,7 @@ export function CannibalisationPanel({ items }: CannibalisationPanelProps) {
                   </li>
                 ))}
               </ul>
+              <AiSuggestionButton request={buildCannibalisationContext(item)} />
             </Card>
           ))}
         </div>
@@ -206,6 +209,7 @@ export function QueryPageMisalignmentPanel({ items }: QueryPageMisalignmentPanel
                 <UrlInspectorButton url={item.suggested_url} />
               </div>
             </div>
+            <AiSuggestionButton request={buildMisalignmentContext(item)} />
           </Card>
         ))}
       </div>
diff --git a/web/src/components/keywordsExplorer/KeywordTabBanner.tsx b/web/src/components/keywordsExplorer/KeywordTabBanner.tsx
index 7903036e..b2c6c83d 100644
--- a/web/src/components/keywordsExplorer/KeywordTabBanner.tsx
+++ b/web/src/components/keywordsExplorer/KeywordTabBanner.tsx
@@ -12,6 +12,9 @@ import {
   Zap,
   Target,
   ArrowRightLeft,
+  Layers,
+  BookOpen,
+  Users,
 } from 'lucide-react';
 import type { KeywordTabId } from './keywordTabMeta';
 import { strings } from '../../lib/strings';
@@ -27,6 +30,9 @@ const TAB_ICONS: Record<KeywordTabId, typeof Key> = {
   cannib: Split,
   alignment: ArrowRightLeft,
   bypage: FileText,
+  topics: Layers,
+  templates: BookOpen,
+  competitor: Users,
 };
 
 export interface KeywordTabBannerProps {
diff --git a/web/src/components/keywordsExplorer/TopicMapPanel.tsx b/web/src/components/keywordsExplorer/TopicMapPanel.tsx
new file mode 100644
index 00000000..0920dd0e
--- /dev/null
+++ b/web/src/components/keywordsExplorer/TopicMapPanel.tsx
@@ -0,0 +1,35 @@
+'use client';
+
+import { Card } from '@/components';
+
+interface ClusterRow {
+  topic?: string;
+  keywords?: string[];
+  size?: number;
+}
+
+interface TopicMapPanelProps {
+  clusters: ClusterRow[];
+  emptyLabel: string;
+}
+
+export default function TopicMapPanel({ clusters, emptyLabel }: TopicMapPanelProps) {
+  if (!clusters.length) {
+    return <p className="text-sm text-muted-foreground py-6">{emptyLabel}</p>;
+  }
+  return (
+    <div className="grid gap-3 sm:grid-cols-2 lg:grid-cols-3">
+      {clusters.slice(0, 24).map((c, i) => (
+        <Card key={i} className="p-4">
+          <h3 className="text-sm font-semibold text-foreground">{c.topic || `Cluster ${i + 1}`}</h3>
+          <p className="text-xs text-muted-foreground mt-1">{c.size ?? c.keywords?.length ?? 0} keywords</p>
+          <ul className="mt-2 text-xs text-foreground space-y-0.5 max-h-28 overflow-y-auto">
+            {(c.keywords || []).slice(0, 8).map((kw) => (
+              <li key={kw} className="truncate">{kw}</li>
+            ))}
+          </ul>
+        </Card>
+      ))}
+    </div>
+  );
+}
diff --git a/web/src/components/keywordsExplorer/keywordTabMeta.ts b/web/src/components/keywordsExplorer/keywordTabMeta.ts
index ad282ef8..d287932a 100644
--- a/web/src/components/keywordsExplorer/keywordTabMeta.ts
+++ b/web/src/components/keywordsExplorer/keywordTabMeta.ts
@@ -17,7 +17,10 @@ export type KeywordTabId =
   | KeywordTableTabId
   | 'cannib'
   | 'alignment'
-  | 'bypage';
+  | 'bypage'
+  | 'topics'
+  | 'templates'
+  | 'competitor';
 
 export function isTableTab(tab: KeywordTabId): tab is KeywordTableTabId {
   return (KEYWORD_TABLE_TAB_IDS as readonly string[]).includes(tab);
diff --git a/web/src/components/lighthouse/DiagnosticItem.tsx b/web/src/components/lighthouse/DiagnosticItem.tsx
index 2d600f51..d2c423fb 100644
--- a/web/src/components/lighthouse/DiagnosticItem.tsx
+++ b/web/src/components/lighthouse/DiagnosticItem.tsx
@@ -1,5 +1,9 @@
+'use client';
+
 import { useState } from 'react';
 import { ChevronDown, ChevronUp } from 'lucide-react';
+import AiSuggestionButton from '@/components/ai/AiSuggestionButton';
+import { buildLighthouseDiagnosticContext } from '@/lib/fixSuggestionContext';
 import type { LighthouseDiagnostic } from '@/types/report';
 import LhDetailsTable from './LhDetailsTable';
 
@@ -53,6 +57,7 @@ export default function DiagnosticItem({ d }: DiagnosticItemProps) {
             {d.detailed_fix && (
               <p className="text-xs text-muted-foreground mt-2">{d.detailed_fix}</p>
             )}
+            <AiSuggestionButton request={buildLighthouseDiagnosticContext(d)} className="mt-3" />
           </div>
 
           {d.estimated_impact && (
diff --git a/web/src/components/lighthouse/LhAuditExpandable.tsx b/web/src/components/lighthouse/LhAuditExpandable.tsx
index 966accc6..6cc85fc9 100644
--- a/web/src/components/lighthouse/LhAuditExpandable.tsx
+++ b/web/src/components/lighthouse/LhAuditExpandable.tsx
@@ -1,5 +1,9 @@
+'use client';
+
 import { useState } from 'react';
 import { ChevronDown, ChevronRight } from 'lucide-react';
+import AiSuggestionButton from '@/components/ai/AiSuggestionButton';
+import { buildLighthouseAuditContext } from '@/lib/fixSuggestionContext';
 import type { LighthouseAuditRef } from '@/types/report';
 import LhDetailsTable from './LhDetailsTable';
 
@@ -40,6 +44,7 @@ export default function LhAuditExpandable({ audit }: LhAuditExpandableProps) {
           )}
           {hasTable && items && <LhDetailsTable headings={headings} items={items} />}
           {!hasTable && <p className="text-xs text-muted-foreground">No detail rows for this audit.</p>}
+          <AiSuggestionButton request={buildLighthouseAuditContext(audit)} />
         </div>
       )}
     </li>
diff --git a/web/src/components/lighthouse/QuickWinCard.tsx b/web/src/components/lighthouse/QuickWinCard.tsx
index f57989d2..e8efa672 100644
--- a/web/src/components/lighthouse/QuickWinCard.tsx
+++ b/web/src/components/lighthouse/QuickWinCard.tsx
@@ -1,5 +1,9 @@
+'use client';
+
 import { useState } from 'react';
 import { CheckCircle, XCircle, ChevronDown, ChevronUp, Zap, Image, Code2, Search, Shield, Clock, type LucideIcon } from 'lucide-react';
+import AiSuggestionButton from '@/components/ai/AiSuggestionButton';
+import { buildLighthouseQuickWinContext } from '@/lib/fixSuggestionContext';
 import type { LighthouseQuickWin } from '@/types/report';
 
 const ICON_MAP: Record<string, LucideIcon> = { Zap, Image, Code2, Search, Shield, Clock };
@@ -59,6 +63,7 @@ export default function QuickWinCard({ win, passed }: QuickWinCardProps) {
             <span className="text-xs text-muted-foreground">Estimated impact:</span>
             <span className="text-xs text-link font-semibold">{win.impact}</span>
           </div>
+          <AiSuggestionButton request={buildLighthouseQuickWinContext(win, passed)} />
         </div>
       )}
     </div>
diff --git a/web/src/components/links/InspectorTabs.tsx b/web/src/components/links/InspectorTabs.tsx
index 8d8256e5..0da6f845 100644
--- a/web/src/components/links/InspectorTabs.tsx
+++ b/web/src/components/links/InspectorTabs.tsx
@@ -13,6 +13,8 @@ import TechnicalTab from './tabs/TechnicalTab';
 import IssuesTab from './tabs/IssuesTab';
 import PageAnalysisTab from './tabs/PageAnalysisTab';
 import SearchRetentionTab from './tabs/SearchRetentionTab';
+import PageImprovePanel from './PageImprovePanel';
+import { Sparkles } from 'lucide-react';
 
 const ci = strings.components.inspectorTabs;
 
@@ -24,9 +26,10 @@ const TAB_ICONS: Record<string, ReactNode> = {
   content: <Code2 className="h-3.5 w-3.5" />,
   technical: <Shield className="h-3.5 w-3.5" />,
   issues: <AlertTriangle className="h-3.5 w-3.5" />,
+  improve: <Sparkles className="h-3.5 w-3.5" />,
 };
 
-const TAB_IDS = ['overview', 'analysis', 'search', 'seo', 'content', 'technical', 'issues'] as const;
+const TAB_IDS = ['overview', 'analysis', 'search', 'seo', 'content', 'technical', 'issues', 'improve'] as const;
 
 const TAB_LABELS: Record<string, string> = {
   overview: ci.overview,
@@ -36,6 +39,7 @@ const TAB_LABELS: Record<string, string> = {
   content: ci.content,
   technical: ci.technical,
   issues: ci.issues,
+  improve: 'Page Improve',
 };
 
 function buildAllIssues(inspectorDetails: InspectorDetails | null): InspectorIssueRow[] {
@@ -148,7 +152,10 @@ export default function InspectorTabs({
         {activeTab === 'content'   && <ContentTab   link={link} />}
         {activeTab === 'technical' && <TechnicalTab  link={link} />}
         {activeTab === 'issues'    && (
-          <IssuesTab lhData={effectiveLh} inspectorDetails={inspectorDetails} />
+          <IssuesTab lhData={effectiveLh} inspectorDetails={inspectorDetails} pageUrl={link.url} />
+        )}
+        {activeTab === 'improve'   && (
+          <PageImprovePanel url={link.url} inspectorDetails={inspectorDetails} />
         )}
       </div>
     </div>
diff --git a/web/src/components/links/LinkAttributesCharts.tsx b/web/src/components/links/LinkAttributesCharts.tsx
new file mode 100644
index 00000000..b6754d5d
--- /dev/null
+++ b/web/src/components/links/LinkAttributesCharts.tsx
@@ -0,0 +1,279 @@
+'use client';
+
+import { useMemo } from 'react';
+import { Bar, Doughnut } from 'react-chartjs-2';
+import type { TooltipItem } from 'chart.js';
+import { Card } from '@/components';
+import { ChartAccessibleFallback } from '@/components/charts';
+import { strings } from '@/lib/strings';
+import type { InlinkAnchorRow, LinkRelSummary } from '@/types/report';
+import { formatPageHrefLines } from '@/utils/linkUtils';
+import { truncateLabel } from '@/components/google/tableUtils';
+import { palette } from '@/utils/chartPalette';
+import { registerChartJsBase, barOptionsHorizontal } from '@/utils/chartJsDefaults';
+import { doughnutOptionsWithPercentTooltip, filterZeroSlices, formatCompositionAria } from '@/lib/chartDoughnutUtils';
+
+registerChartJsBase();
+
+const TOP_N = 10;
+
+function aggregateInlinks(
+  rows: InlinkAnchorRow[],
+  key: 'anchor_text' | 'target_url',
+  limit = TOP_N,
+): Array<{ label: string; value: number; title?: string }> {
+  const totals = new Map<string, { value: number; title?: string }>();
+  for (const row of rows) {
+    const raw = (row[key] ?? '').trim();
+    const mapKey = raw || '(empty)';
+    const prev = totals.get(mapKey)?.value ?? 0;
+    const title = key === 'target_url' ? raw || mapKey : undefined;
+    totals.set(mapKey, {
+      value: prev + (row.inlink_count ?? 0),
+      title,
+    });
+  }
+  return [...totals.entries()]
+    .map(([label, { value, title }]) => ({
+      label: key === 'target_url' ? formatPageHrefLines(title ?? label).label : label,
+      value,
+      title: title ?? label,
+    }))
+    .sort((a, b) => b.value - a.value)
+    .slice(0, limit);
+}
+
+function linkBarOptions(titleAtIndex: (index: number) => string) {
+  const base = barOptionsHorizontal();
+  return {
+    ...base,
+    plugins: {
+      ...base.plugins,
+      tooltip: {
+        callbacks: {
+          title: (items: TooltipItem<'bar'>[]) => {
+            const idx = items[0]?.dataIndex ?? 0;
+            return titleAtIndex(idx) || items[0]?.label || '';
+          },
+          label: (ctx: TooltipItem<'bar'>) => ` ${Number(ctx.raw).toLocaleString()} inlinks`,
+        },
+      },
+    },
+    scales: {
+      ...base.scales,
+      y: {
+        ...(base.scales?.y as object),
+        ticks: {
+          maxWidth: 112,
+          font: { size: 11 },
+          callback(this: { getLabelForValue: (v: number) => string }, value: string | number) {
+            const label = this.getLabelForValue(Number(value));
+            return truncateLabel(label, 36);
+          },
+        },
+      },
+    },
+  };
+}
+
+interface LinkAttributesChartsProps {
+  summary?: LinkRelSummary | null;
+  anchors?: InlinkAnchorRow[];
+  labels: {
+    internal: string;
+    external: string;
+    nofollow: string;
+    sponsored: string;
+    follow: string;
+    ugc: string;
+  };
+}
+
+export default function LinkAttributesCharts({ summary, anchors, labels }: LinkAttributesChartsProps) {
+  const vl = strings.views.links;
+
+  const scopeChart = useMemo(() => {
+    if (!summary) return null;
+    const internal = summary.internal_edges ?? 0;
+    const external = summary.external_edges ?? 0;
+    const { labels: sliceLabels, values } = filterZeroSlices(
+      [labels.internal, labels.external],
+      [internal, external],
+    );
+    if (values.length === 0) return null;
+    return {
+      labels: sliceLabels,
+      values,
+      colors: palette(values.length),
+      aria: formatCompositionAria(sliceLabels, values, 'links'),
+    };
+  }, [summary, labels.internal, labels.external]);
+
+  const internalAttrsChart = useMemo(() => {
+    if (!summary) return null;
+    const internal = summary.internal_edges ?? 0;
+    const nofollow = summary.nofollow_internal ?? 0;
+    const sponsored = summary.sponsored_internal ?? 0;
+    const ugc = summary.ugc_internal ?? 0;
+    const follow = Math.max(0, internal - nofollow - sponsored - ugc);
+    const sliceLabels = [labels.follow, labels.nofollow, labels.sponsored, labels.ugc];
+    const rawValues = [follow, nofollow, sponsored, ugc];
+    const { labels: filteredLabels, values } = filterZeroSlices(sliceLabels, rawValues);
+    if (values.length === 0) return null;
+    return {
+      labels: filteredLabels,
+      values,
+      colors: palette(values.length),
+      aria: formatCompositionAria(filteredLabels, values, 'internal links'),
+    };
+  }, [summary, labels.follow, labels.nofollow, labels.sponsored, labels.ugc]);
+
+  const topAnchors = useMemo(
+    () => (anchors?.length ? aggregateInlinks(anchors, 'anchor_text') : []),
+    [anchors],
+  );
+
+  const topTargets = useMemo(
+    () => (anchors?.length ? aggregateInlinks(anchors, 'target_url') : []),
+    [anchors],
+  );
+
+  const anchorBarOpts = useMemo(
+    () => linkBarOptions((idx) => topAnchors[idx]?.label ?? ''),
+    [topAnchors],
+  );
+
+  const targetBarOpts = useMemo(
+    () => linkBarOptions((idx) => topTargets[idx]?.title ?? topTargets[idx]?.label ?? ''),
+    [topTargets],
+  );
+
+  const hasCharts = scopeChart || internalAttrsChart || topAnchors.length > 0 || topTargets.length > 0;
+  if (!hasCharts) return null;
+
+  return (
+    <div className="grid grid-cols-1 lg:grid-cols-2 gap-4 min-w-0">
+      {scopeChart ? (
+        <Card padding="tight" shadow className="min-w-0 overflow-hidden">
+          <h2 className="text-sm font-bold text-foreground mb-1">{vl.chartLinkScopeTitle}</h2>
+          <p className="text-xs text-muted-foreground mb-3">{vl.chartLinkScopeHint}</p>
+          <div className="h-56 flex items-center justify-center">
+            <div className="w-full max-w-[260px] h-48">
+              <ChartAccessibleFallback
+                summary={scopeChart.aria}
+                rows={scopeChart.labels.map((label, i) => [label, scopeChart.values[i] ?? 0])}
+              >
+                <Doughnut
+                  data={{
+                    labels: scopeChart.labels,
+                    datasets: [
+                      {
+                        data: scopeChart.values,
+                        backgroundColor: scopeChart.colors,
+                        borderColor: 'rgba(15,23,42,0.8)',
+                        borderWidth: 2,
+                      },
+                    ],
+                  }}
+                  options={doughnutOptionsWithPercentTooltip()}
+                />
+              </ChartAccessibleFallback>
+            </div>
+          </div>
+        </Card>
+      ) : null}
+
+      {internalAttrsChart ? (
+        <Card padding="tight" shadow className="min-w-0 overflow-hidden">
+          <h2 className="text-sm font-bold text-foreground mb-1">{vl.chartInternalAttrsTitle}</h2>
+          <p className="text-xs text-muted-foreground mb-3">{vl.chartInternalAttrsHint}</p>
+          <div className="h-56 flex items-center justify-center">
+            <div className="w-full max-w-[260px] h-48">
+              <ChartAccessibleFallback
+                summary={internalAttrsChart.aria}
+                rows={internalAttrsChart.labels.map((label, i) => [label, internalAttrsChart.values[i] ?? 0])}
+              >
+                <Doughnut
+                  data={{
+                    labels: internalAttrsChart.labels,
+                    datasets: [
+                      {
+                        data: internalAttrsChart.values,
+                        backgroundColor: internalAttrsChart.colors,
+                        borderColor: 'rgba(15,23,42,0.8)',
+                        borderWidth: 2,
+                      },
+                    ],
+                  }}
+                  options={doughnutOptionsWithPercentTooltip()}
+                />
+              </ChartAccessibleFallback>
+            </div>
+          </div>
+        </Card>
+      ) : null}
+
+      {topAnchors.length > 0 ? (
+        <Card padding="tight" shadow className="min-w-0 overflow-hidden">
+          <h2 className="text-sm font-bold text-foreground mb-1">{vl.chartTopAnchorsTitle}</h2>
+          <p className="text-xs text-muted-foreground mb-3">{vl.chartTopAnchorsHint}</p>
+          <div className="relative h-72 min-w-0 w-full overflow-hidden">
+            <ChartAccessibleFallback
+              summary={formatCompositionAria(
+                topAnchors.map((r) => r.label),
+                topAnchors.map((r) => r.value),
+                'inlinks',
+              )}
+              rows={topAnchors.map((r) => [r.label, r.value])}
+            >
+              <Bar
+                data={{
+                  labels: topAnchors.map((r) => truncateLabel(r.label, 36)),
+                  datasets: [
+                    {
+                      data: topAnchors.map((r) => r.value),
+                      backgroundColor: palette(topAnchors.length),
+                      borderRadius: 4,
+                    },
+                  ],
+                }}
+                options={anchorBarOpts}
+              />
+            </ChartAccessibleFallback>
+          </div>
+        </Card>
+      ) : null}
+
+      {topTargets.length > 0 ? (
+        <Card padding="tight" shadow className="min-w-0 overflow-hidden">
+          <h2 className="text-sm font-bold text-foreground mb-1">{vl.chartTopTargetsTitle}</h2>
+          <p className="text-xs text-muted-foreground mb-3">{vl.chartTopTargetsHint}</p>
+          <div className="relative h-72 min-w-0 w-full overflow-hidden">
+            <ChartAccessibleFallback
+              summary={formatCompositionAria(
+                topTargets.map((r) => r.title ?? r.label),
+                topTargets.map((r) => r.value),
+                'inlinks',
+              )}
+              rows={topTargets.map((r) => [r.title ?? r.label, r.value])}
+            >
+              <Bar
+                data={{
+                  labels: topTargets.map((r) => truncateLabel(r.label, 36)),
+                  datasets: [
+                    {
+                      data: topTargets.map((r) => r.value),
+                      backgroundColor: palette(topTargets.length),
+                      borderRadius: 4,
+                    },
+                  ],
+                }}
+                options={targetBarOpts}
+              />
+            </ChartAccessibleFallback>
+          </div>
+        </Card>
+      ) : null}
+    </div>
+  );
+}
diff --git a/web/src/components/links/LinkAttributesPanel.tsx b/web/src/components/links/LinkAttributesPanel.tsx
new file mode 100644
index 00000000..924e5da3
--- /dev/null
+++ b/web/src/components/links/LinkAttributesPanel.tsx
@@ -0,0 +1,88 @@
+'use client';
+
+import SortablePaginatedTable from '@/components/google/SortablePaginatedTable';
+import { Card } from '@/components';
+import { strings } from '@/lib/strings';
+import type { InlinkAnchorRow, LinkRelSummary } from '@/types/report';
+import type { TableColumn } from '@/types/components';
+import LinkAttributesCharts from './LinkAttributesCharts';
+
+const paginationLabels = {
+  showingSlice: strings.views.backlinks.table.showingSlice,
+  pageOf: strings.views.backlinks.table.pageOf,
+  of: strings.views.backlinks.table.of,
+  previous: strings.views.backlinks.table.previous,
+  next: strings.views.backlinks.table.next,
+  rowsPerPage: strings.views.backlinks.table.rowsPerPage,
+};
+
+interface LinkAttributesPanelProps {
+  summary?: LinkRelSummary | null;
+  anchors?: InlinkAnchorRow[];
+  labels: {
+    title: string;
+    total: string;
+    internal: string;
+    nofollow: string;
+    sponsored: string;
+    external: string;
+    anchorMatrix: string;
+    target: string;
+    anchor: string;
+    inlinks: string;
+    follow: string;
+    ugc: string;
+  };
+}
+
+export default function LinkAttributesPanel({ summary, anchors, labels }: LinkAttributesPanelProps) {
+  if (!summary && !(anchors?.length)) return null;
+
+  const columns: TableColumn[] = [
+    { key: 'target_url', label: labels.target },
+    { key: 'anchor_text', label: labels.anchor },
+    {
+      key: 'inlink_count',
+      label: labels.inlinks,
+      render: (v) => (typeof v === 'number' ? v.toLocaleString() : String(v ?? '')),
+    },
+  ];
+
+  return (
+    <div className="space-y-4 min-w-0">
+      <LinkAttributesCharts
+        summary={summary}
+        anchors={anchors}
+        labels={{
+          internal: labels.internal,
+          external: labels.external,
+          nofollow: labels.nofollow,
+          sponsored: labels.sponsored,
+          follow: labels.follow,
+          ugc: labels.ugc,
+        }}
+      />
+      {summary ? (
+        <Card className="p-4 grid grid-cols-2 sm:grid-cols-3 lg:grid-cols-6 gap-3 text-sm">
+          <div><span className="text-muted-foreground">{labels.total}</span><div className="font-semibold">{(summary.total_edges ?? 0).toLocaleString()}</div></div>
+          <div><span className="text-muted-foreground">{labels.internal}</span><div className="font-semibold">{(summary.internal_edges ?? 0).toLocaleString()}</div></div>
+          <div><span className="text-muted-foreground">{labels.nofollow}</span><div className="font-semibold">{(summary.nofollow_internal ?? 0).toLocaleString()}</div></div>
+          <div><span className="text-muted-foreground">{labels.sponsored}</span><div className="font-semibold">{(summary.sponsored_internal ?? 0).toLocaleString()}</div></div>
+          <div><span className="text-muted-foreground">{labels.external}</span><div className="font-semibold">{(summary.external_edges ?? 0).toLocaleString()}</div></div>
+        </Card>
+      ) : null}
+      {anchors?.length ? (
+        <Card className="p-4 min-w-0 overflow-hidden">
+          <h3 className="text-sm font-semibold mb-3">{labels.anchorMatrix}</h3>
+          <SortablePaginatedTable
+            rows={anchors as Record<string, unknown>[]}
+            columns={columns}
+            defaultSort="inlink_count"
+            defaultDir="desc"
+            paginationLabels={paginationLabels}
+          />
+        </Card>
+      ) : null}
+    </div>
+  );
+}
diff --git a/web/src/components/links/PageImprovePanel.tsx b/web/src/components/links/PageImprovePanel.tsx
new file mode 100644
index 00000000..94e7e7c3
--- /dev/null
+++ b/web/src/components/links/PageImprovePanel.tsx
@@ -0,0 +1,116 @@
+'use client';
+
+import { useCallback, useState } from 'react';
+import { Loader2, Sparkles } from 'lucide-react';
+import { apiUrl } from '@/lib/publicBase';
+import { strings } from '@/lib/strings';
+import { useOptionalReport } from '@/context/useReport';
+import { useReadOnlySession } from '@/hooks/useReadOnlySession';
+import type { InspectorDetails } from '@/types/report';
+import { Button } from '@/components';
+
+const FIX_TEMPLATES: Record<string, string> = {
+  title: 'Add a unique, descriptive title tag (50–60 characters) with primary keyword near the start.',
+  meta: 'Write a compelling meta description (120–160 characters) that matches search intent.',
+  canonical: 'Set rel=canonical to the preferred URL version of this page.',
+  h1: 'Use exactly one H1 that describes the main topic; align it with the title where appropriate.',
+  noindex: 'Remove noindex if this page should rank; keep it only for thin or duplicate URLs.',
+  broken: 'Fix or remove links to broken URLs; update redirects at the source.',
+  accessibility: 'Address axe/Lighthouse accessibility findings on this URL.',
+};
+
+interface PageCoachResult {
+  summary?: string;
+  actions?: string[];
+  provenance?: string;
+}
+
+interface PageImprovePanelProps {
+  url: string;
+  inspectorDetails: InspectorDetails | null;
+}
+
+export default function PageImprovePanel({ url, inspectorDetails }: PageImprovePanelProps) {
+  const pi = strings.components.pageImprove;
+  const reportCtx = useOptionalReport();
+  const { readOnly } = useReadOnlySession();
+  const [coachLoading, setCoachLoading] = useState(false);
+  const [coachError, setCoachError] = useState<string | null>(null);
+  const [coach, setCoach] = useState<PageCoachResult | null>(null);
+
+  const items = inspectorDetails?.categoryIssues || [];
+  const checklist = items.length
+    ? items.map((iss) => ({
+        message: iss.message || 'Issue',
+        recommendation: iss.recommendation || FIX_TEMPLATES.accessibility,
+        priority: iss.priority || 'Medium',
+      }))
+    : [
+        { message: 'No open category issues for this URL.', recommendation: FIX_TEMPLATES.title, priority: 'Low' },
+      ];
+
+  const fetchCoach = useCallback(async () => {
+    if (readOnly) return;
+    setCoachLoading(true);
+    setCoachError(null);
+    try {
+      const body: Record<string, unknown> = { url };
+      const reportId = reportCtx?.selectedReportId;
+      if (reportId != null) {
+        body.currentType = 'snapshot';
+        body.currentId = reportId;
+      }
+      const res = await fetch(apiUrl('/links/page-coach'), {
+        method: 'POST',
+        headers: { 'Content-Type': 'application/json' },
+        body: JSON.stringify(body),
+      });
+      const payload = await res.json();
+      if (!res.ok || !payload.ok) {
+        throw new Error(payload.error || pi.coachFailed);
+      }
+      setCoach((payload.coach || null) as PageCoachResult | null);
+    } catch (e) {
+      setCoachError(e instanceof Error ? e.message : pi.coachFailed);
+    } finally {
+      setCoachLoading(false);
+    }
+  }, [readOnly, url, reportCtx?.selectedReportId, pi.coachFailed]);
+
+  return (
+    <div className="space-y-4 p-4">
+      <p className="text-sm text-muted-foreground">
+        Page Improve checklist for <span className="font-mono text-foreground break-all">{url}</span>
+      </p>
+      {!readOnly ? (
+        <Button type="button" variant="secondary" className="!text-xs" onClick={() => void fetchCoach()} disabled={coachLoading}>
+          {coachLoading ? <Loader2 className="h-3.5 w-3.5 animate-spin" aria-hidden /> : <Sparkles className="h-3.5 w-3.5" aria-hidden />}
+          {coachLoading ? pi.coachLoading : pi.coachButton}
+        </Button>
+      ) : null}
+      {coachError ? <p className="text-xs text-red-700 dark:text-red-400">{coachError}</p> : null}
+      {coach?.summary ? (
+        <div className="rounded-lg border border-default bg-brand-900/40 p-3 space-y-2">
+          <p className="text-xs font-semibold text-foreground">{pi.coachTitle}</p>
+          <pre className="whitespace-pre-wrap text-xs text-muted-foreground font-sans leading-relaxed">{coach.summary}</pre>
+          {coach.actions?.length ? (
+            <ul className="list-disc pl-4 text-xs text-muted-foreground space-y-1">
+              {coach.actions.slice(0, 8).map((action) => (
+                <li key={action}>{action}</li>
+              ))}
+            </ul>
+          ) : null}
+        </div>
+      ) : null}
+      <ol className="space-y-3 list-decimal list-inside">
+        {checklist.slice(0, 12).map((item, i) => (
+          <li key={i} className="text-sm">
+            <span className="font-medium text-foreground">{item.message}</span>
+            <span className="ml-2 text-xs rounded px-1.5 py-0.5 bg-brand-800 text-muted-foreground">{item.priority}</span>
+            <p className="text-xs text-muted-foreground mt-1 ml-5">{item.recommendation}</p>
+          </li>
+        ))}
+      </ol>
+    </div>
+  );
+}
diff --git a/web/src/components/links/SavedCrawlFiltersBar.tsx b/web/src/components/links/SavedCrawlFiltersBar.tsx
new file mode 100644
index 00000000..7582a97e
--- /dev/null
+++ b/web/src/components/links/SavedCrawlFiltersBar.tsx
@@ -0,0 +1,96 @@
+'use client';
+
+import { useCallback, useEffect, useState } from 'react';
+import { BookmarkPlus } from 'lucide-react';
+import { Button } from '@/components';
+import { apiUrl } from '@/lib/publicBase';
+import type { LinksFilterValues } from './LinksFilterBar';
+
+interface SavedCrawlFiltersBarProps {
+  propertyId: number;
+  filterValues: LinksFilterValues;
+  onLoad: (values: LinksFilterValues) => void;
+}
+
+export default function SavedCrawlFiltersBar({ propertyId, filterValues, onLoad }: SavedCrawlFiltersBarProps) {
+  const [names, setNames] = useState<string[]>([]);
+  const [selected, setSelected] = useState('');
+  const [status, setStatus] = useState('');
+
+  const refresh = useCallback(async () => {
+    if (!propertyId) {
+      setNames([]);
+      return;
+    }
+    try {
+      const res = await fetch(apiUrl(`/filters?propertyId=${propertyId}`));
+      const body = await res.json();
+      setNames((body.filters || []).map((f: { name: string }) => f.name));
+    } catch {
+      setNames([]);
+    }
+  }, [propertyId]);
+
+  useEffect(() => {
+    void refresh();
+  }, [refresh]);
+
+  const save = async () => {
+    const name = window.prompt('Filter name');
+    if (!name?.trim() || !propertyId) return;
+    setStatus('');
+    try {
+      const res = await fetch(apiUrl('/filters'), {
+        method: 'POST',
+        headers: { 'Content-Type': 'application/json' },
+        body: JSON.stringify({ propertyId, name: name.trim(), filterJson: filterValues }),
+      });
+      if (!res.ok) throw new Error('Save failed');
+      setStatus(`Saved "${name.trim()}"`);
+      await refresh();
+    } catch {
+      setStatus('Could not save filter');
+    }
+  };
+
+  const load = async () => {
+    if (!selected || !propertyId) return;
+    try {
+      const res = await fetch(apiUrl(`/filters?propertyId=${propertyId}`));
+      const body = await res.json();
+      const row = (body.filters || []).find((f: { name: string }) => f.name === selected);
+      if (row?.filterJson) onLoad(row.filterJson as LinksFilterValues);
+    } catch {
+      setStatus('Could not load filter');
+    }
+  };
+
+  if (!propertyId) return null;
+
+  return (
+    <div className="flex flex-wrap items-center gap-2 text-xs">
+      <Button type="button" variant="secondary" className="!py-1 !px-2 !text-xs" onClick={() => void save()}>
+        <BookmarkPlus className="h-3.5 w-3.5" />
+        Save filter
+      </Button>
+      {names.length ? (
+        <>
+          <select
+            value={selected}
+            onChange={(e) => setSelected(e.target.value)}
+            className="rounded border border-default bg-background px-2 py-1 text-xs"
+          >
+            <option value="">Load saved…</option>
+            {names.map((n) => (
+              <option key={n} value={n}>{n}</option>
+            ))}
+          </select>
+          <Button type="button" variant="secondary" className="!py-1 !px-2 !text-xs" onClick={() => void load()} disabled={!selected}>
+            Load
+          </Button>
+        </>
+      ) : null}
+      {status ? <span className="text-muted-foreground">{status}</span> : null}
+    </div>
+  );
+}
diff --git a/web/src/components/links/SecHeaderRow.tsx b/web/src/components/links/SecHeaderRow.tsx
index ef193230..8b7da04b 100644
--- a/web/src/components/links/SecHeaderRow.tsx
+++ b/web/src/components/links/SecHeaderRow.tsx
@@ -1,13 +1,18 @@
+'use client';
+
 import { useState } from 'react';
 import { CheckCircle, XCircle, ChevronDown, ChevronUp } from 'lucide-react';
+import AiSuggestionButton from '@/components/ai/AiSuggestionButton';
+import { buildSecurityHeaderContext } from '@/lib/fixSuggestionContext';
 
 export interface SecHeaderRowProps {
   label: string;
   value?: string | null;
   recommendation?: string;
+  pageUrl?: string;
 }
 
-export default function SecHeaderRow({ label, value, recommendation }: SecHeaderRowProps) {
+export default function SecHeaderRow({ label, value, recommendation, pageUrl }: SecHeaderRowProps) {
   const [open, setOpen] = useState(false);
   const present = !!value;
 
@@ -43,6 +48,9 @@ export default function SecHeaderRow({ label, value, recommendation }: SecHeader
               <span className="text-muted-foreground">Recommendation:</span> {recommendation}
             </p>
           )}
+          {!present ? (
+            <AiSuggestionButton request={buildSecurityHeaderContext(label, pageUrl)} />
+          ) : null}
         </div>
       )}
     </div>
diff --git a/web/src/components/links/SerpPreview.tsx b/web/src/components/links/SerpPreview.tsx
new file mode 100644
index 00000000..1f78b39d
--- /dev/null
+++ b/web/src/components/links/SerpPreview.tsx
@@ -0,0 +1,31 @@
+'use client';
+
+import type { LinkDetail } from '@/types/report';
+
+interface SerpPreviewProps {
+  link: LinkDetail;
+  domain?: string;
+}
+
+export default function SerpPreview({ link, domain }: SerpPreviewProps) {
+  const title = (link.title || 'Missing title').slice(0, 60);
+  const desc = (link.meta_description || 'Missing meta description').slice(0, 160);
+  const url = link.canonical_url || link.url;
+  let displayUrl = url;
+  try {
+    const u = new URL(url);
+    displayUrl = domain || `${u.hostname}${u.pathname}`.slice(0, 80);
+  } catch {
+    displayUrl = url.slice(0, 80);
+  }
+
+  return (
+    <div className="rounded-xl border border-default bg-white dark:bg-brand-950 p-4 max-w-xl">
+      <p className="text-[#1a0dab] dark:text-blue-400 text-lg leading-snug hover:underline cursor-default truncate">
+        {title}
+      </p>
+      <p className="text-sm text-[#006621] dark:text-emerald-500 truncate mt-0.5">{displayUrl}</p>
+      <p className="text-sm text-[#4d5156] dark:text-muted-foreground mt-1 line-clamp-2">{desc}</p>
+    </div>
+  );
+}
diff --git a/web/src/components/links/explorer/LinksExplorerAnchorsTab.tsx b/web/src/components/links/explorer/LinksExplorerAnchorsTab.tsx
new file mode 100644
index 00000000..7d21bd9f
--- /dev/null
+++ b/web/src/components/links/explorer/LinksExplorerAnchorsTab.tsx
@@ -0,0 +1,32 @@
+'use client';
+
+import LinkAttributesPanel from '@/components/links/LinkAttributesPanel';
+import type { InlinkAnchorRow, LinkRelSummary } from '@/types/report';
+import { LinksExplorerTabPanel } from './LinksExplorerTabPanel';
+
+export interface LinksExplorerAnchorsTabProps {
+  summary?: LinkRelSummary | null;
+  anchors?: InlinkAnchorRow[];
+  labels: {
+    title: string;
+    total: string;
+    internal: string;
+    nofollow: string;
+    sponsored: string;
+    external: string;
+    anchorMatrix: string;
+    target: string;
+    anchor: string;
+    inlinks: string;
+    follow: string;
+    ugc: string;
+  };
+}
+
+export function LinksExplorerAnchorsTab({ summary, anchors, labels }: LinksExplorerAnchorsTabProps) {
+  return (
+    <LinksExplorerTabPanel tabId="anchors" className="flex flex-col gap-4 min-w-0">
+      <LinkAttributesPanel summary={summary} anchors={anchors} labels={labels} />
+    </LinksExplorerTabPanel>
+  );
+}
diff --git a/web/src/components/links/explorer/LinksExplorerTabPanel.tsx b/web/src/components/links/explorer/LinksExplorerTabPanel.tsx
index c5571c98..a4eb45bc 100644
--- a/web/src/components/links/explorer/LinksExplorerTabPanel.tsx
+++ b/web/src/components/links/explorer/LinksExplorerTabPanel.tsx
@@ -1,7 +1,9 @@
 import type { ReactNode } from 'react';
 
+export type LinksExplorerTabId = 'urls' | 'anchors';
+
 interface LinksExplorerTabPanelProps {
-  tabId: 'urls';
+  tabId: LinksExplorerTabId;
   className?: string;
   children: ReactNode;
 }
diff --git a/web/src/components/links/explorer/LinksExplorerTableTab.tsx b/web/src/components/links/explorer/LinksExplorerTableTab.tsx
index ad8a7f92..b8e2f78d 100644
--- a/web/src/components/links/explorer/LinksExplorerTableTab.tsx
+++ b/web/src/components/links/explorer/LinksExplorerTableTab.tsx
@@ -7,7 +7,9 @@ import { strings, format } from '@/lib/strings';
 import { Card, Badge, Button } from '@/components';
 import { formatMs, rtColor, formatPageHrefLines } from '@/utils/linkUtils';
 import { linkHasBrowserErrors } from '@/lib/browserErrors';
+import { collectCustomFieldKeys, parseLinkCustomFields } from '@/lib/customFields';
 import { SortTh, RowTooltip, LinksFilterBar, InlinksMetricCell } from '@/components/links';
+import SavedCrawlFiltersBar from '@/components/links/SavedCrawlFiltersBar';
 import type { LinksFilterValues } from '@/components/links/LinksFilterBar';
 import type { LinkSortKey } from './types';
 import { LinksExplorerTabPanel } from './LinksExplorerTabPanel';
@@ -16,6 +18,8 @@ export interface LinksExplorerTableTabProps {
   filterValues: LinksFilterValues;
   onFilterChange: (key: keyof LinksFilterValues, value: string) => void;
   onClearAllFilters: () => void;
+  propertyId?: number;
+  onLoadSavedFilter?: (values: LinksFilterValues) => void;
   searchQuery: string;
   filtered: ReportLink[];
   pageLinks: ReportLink[];
@@ -40,6 +44,8 @@ export function LinksExplorerTableTab({
   filterValues,
   onFilterChange,
   onClearAllFilters,
+  propertyId = 0,
+  onLoadSavedFilter,
   searchQuery,
   filtered,
   pageLinks,
@@ -62,6 +68,7 @@ export function LinksExplorerTableTab({
   const vl = strings.views.links;
   const sj = strings.common;
   const hasCustomExtract = links.some((l) => l.custom_extract);
+  const customFieldKeys = collectCustomFieldKeys(links);
 
   return (
     <LinksExplorerTabPanel tabId="urls" className="flex flex-col gap-4">
@@ -71,6 +78,13 @@ export function LinksExplorerTableTab({
         onClearAll={onClearAllFilters}
         searchQuery={searchQuery}
       />
+      {onLoadSavedFilter ? (
+        <SavedCrawlFiltersBar
+          propertyId={propertyId}
+          filterValues={filterValues}
+          onLoad={onLoadSavedFilter}
+        />
+      ) : null}
 
       <Card overflowHidden padding="none" className="flex flex-col min-h-[min(500px,70vh)] sm:min-h-[500px]">
         <div
@@ -123,6 +137,14 @@ export function LinksExplorerTableTab({
                     {vl.thCustomExtract}
                   </th>
                 ) : null}
+                {customFieldKeys.map((key) => (
+                  <th
+                    key={key}
+                    className="hidden xl:table-cell px-4 py-4 text-muted-foreground uppercase text-xs whitespace-nowrap"
+                  >
+                    {key}
+                  </th>
+                ))}
                 <th className="hidden lg:table-cell px-4 py-4 text-muted-foreground uppercase text-xs whitespace-nowrap">
                   {vl.thJsErrors}
                 </th>
@@ -207,6 +229,18 @@ export function LinksExplorerTableTab({
                         {link.custom_extract || sj.emDash}
                       </td>
                     ) : null}
+                    {customFieldKeys.map((key) => {
+                      const value = parseLinkCustomFields(link)[key];
+                      return (
+                        <td
+                          key={key}
+                          className="hidden xl:table-cell px-4 py-3 text-xs text-foreground align-middle max-w-[10rem] truncate"
+                          title={value}
+                        >
+                          {value || sj.emDash}
+                        </td>
+                      );
+                    })}
                     <td className="hidden lg:table-cell px-4 py-3 text-xs align-middle whitespace-nowrap">
                       {linkHasBrowserErrors(link) ? (
                         <span className="inline-flex items-center rounded-md bg-red-500/10 border border-red-500/25 px-2 py-0.5 font-mono text-red-700 dark:text-red-300">
diff --git a/web/src/components/links/explorer/index.ts b/web/src/components/links/explorer/index.ts
index 7303b6b8..c3ce7477 100644
--- a/web/src/components/links/explorer/index.ts
+++ b/web/src/components/links/explorer/index.ts
@@ -1,3 +1,5 @@
 export type { LinkSortKey } from './types';
+export type { LinksExplorerTabId } from './LinksExplorerTabPanel';
+export { LinksExplorerAnchorsTab } from './LinksExplorerAnchorsTab';
 export { LinksExplorerTabPanel } from './LinksExplorerTabPanel';
 export { LinksExplorerTableTab } from './LinksExplorerTableTab';
diff --git a/web/src/components/links/tabs/ContentTab.tsx b/web/src/components/links/tabs/ContentTab.tsx
index ab9d5f56..ed110c80 100644
--- a/web/src/components/links/tabs/ContentTab.tsx
+++ b/web/src/components/links/tabs/ContentTab.tsx
@@ -11,6 +11,9 @@ import { PALETTE_CATEGORICAL } from '../../../utils/chartPalette';
 import HeadingPills from '../HeadingPills';
 import { strings, format } from '../../../lib/strings';
 import { getGridColor, getChartTitleColor } from '../../../utils/chartJsDefaults';
+import AiSuggestionButton from '@/components/ai/AiSuggestionButton';
+import { buildContentSignalContext } from '@/lib/fixSuggestionContext';
+import type { FixSuggestionRequest } from '@/types/fixSuggestion';
 
 ChartJS.register(CategoryScale, LinearScale, BarElement, Title, Tooltip, Legend);
 
@@ -118,16 +121,29 @@ function SectionHeader({ icon, title, description }: SectionHeaderProps) {
   );
 }
 
-function QualityStatusRow({ label, detail, statusLabel }: { label: string; detail: string; statusLabel: string }) {
+function QualityStatusRow({
+  label,
+  detail,
+  statusLabel,
+  fixRequest,
+}: {
+  label: string;
+  detail: string;
+  statusLabel: string;
+  fixRequest?: FixSuggestionRequest;
+}) {
   return (
-    <div className="flex flex-col gap-2 sm:flex-row sm:items-center sm:justify-between">
-      <div>
-        <div className="text-sm font-bold text-foreground">{label}</div>
-        <div className="text-xs text-muted-foreground">{detail}</div>
+    <div className="space-y-2">
+      <div className="flex flex-col gap-2 sm:flex-row sm:items-center sm:justify-between">
+        <div>
+          <div className="text-sm font-bold text-foreground">{label}</div>
+          <div className="text-xs text-muted-foreground">{detail}</div>
+        </div>
+        <span className={`inline-flex w-fit items-center rounded-full border px-2.5 py-1 text-xs font-semibold ${qualityBadgeClass(statusLabel)}`}>
+          {statusLabel}
+        </span>
       </div>
-      <span className={`inline-flex w-fit items-center rounded-full border px-2.5 py-1 text-xs font-semibold ${qualityBadgeClass(statusLabel)}`}>
-        {statusLabel}
-      </span>
+      {fixRequest ? <AiSuggestionButton request={fixRequest} /> : null}
     </div>
   );
 }
@@ -177,9 +193,13 @@ export default function ContentTab({ link }: ContentTabProps) {
   const metaQualLabels = vca.metaQualLabels;
   const h1QualLabels = vca.h1Labels;
 
-  const titleStatus = titleQualLabels[titleQualityIndex(titleLen)] ?? sj.emDash;
-  const metaStatus = metaQualLabels[metaQualityIndex(metaLen)] ?? sj.emDash;
-  const h1Status = h1c != null && !Number.isNaN(h1c) ? h1QualLabels[h1QualityIndex(h1c)] : null;
+  const titleQi = titleQualityIndex(titleLen);
+  const metaQi = metaQualityIndex(metaLen);
+  const h1Qi = h1c != null && !Number.isNaN(h1c) ? h1QualityIndex(h1c) : null;
+  const titleStatus = titleQualLabels[titleQi] ?? sj.emDash;
+  const metaStatus = metaQualLabels[metaQi] ?? sj.emDash;
+  const h1Status = h1Qi != null ? h1QualLabels[h1Qi] : null;
+  const pageUrl = link.url || '';
 
   const kwSorted = useMemo(() => {
     const rows = keywords.map((kw) => normaliseKw(kw)).filter((k) => k.word);
@@ -332,6 +352,11 @@ export default function ContentTab({ link }: ContentTabProps) {
               label={lc.titleTag}
               detail={format(lc.characters, { n: titleLen })}
               statusLabel={titleStatus}
+              fixRequest={
+                titleQi !== 2
+                  ? buildContentSignalContext('title', lc.titleTag, format(lc.characters, { n: titleLen }), titleStatus, pageUrl)
+                  : undefined
+              }
             />
           </div>
           <div className="py-4 first:pt-0 last:pb-0">
@@ -339,6 +364,11 @@ export default function ContentTab({ link }: ContentTabProps) {
               label={lc.metaDesc}
               detail={format(lc.characters, { n: metaLen })}
               statusLabel={metaStatus}
+              fixRequest={
+                metaQi !== 2
+                  ? buildContentSignalContext('meta', lc.metaDesc, format(lc.characters, { n: metaLen }), metaStatus, pageUrl)
+                  : undefined
+              }
             />
           </div>
           <div className="py-4 first:pt-0 last:pb-0">
@@ -347,6 +377,11 @@ export default function ContentTab({ link }: ContentTabProps) {
                 label={lc.h1Count}
                 detail={format(lc.headingCount, { n: h1c ?? 0 })}
                 statusLabel={h1Status}
+                fixRequest={
+                  h1Qi !== 1
+                    ? buildContentSignalContext('h1', lc.h1Count, format(lc.headingCount, { n: h1c ?? 0 }), h1Status, pageUrl)
+                    : undefined
+                }
               />
             ) : (
               <div className="text-sm text-muted-foreground">{lc.noH1Data}</div>
diff --git a/web/src/components/links/tabs/IssuesTab.tsx b/web/src/components/links/tabs/IssuesTab.tsx
index 78377ca8..89d8e345 100644
--- a/web/src/components/links/tabs/IssuesTab.tsx
+++ b/web/src/components/links/tabs/IssuesTab.tsx
@@ -9,15 +9,22 @@ import { palette, scoreBandColor } from '../../../utils/chartPalette';
 import { registerChartJsBase, barOptionsHorizontal } from '../../../utils/chartJsDefaults';
 import { RankedBarChart } from '../../../components/charts';
 import { formatCompositionAria } from '../../../lib/chartDoughnutUtils';
+import AiSuggestionButton from '@/components/ai/AiSuggestionButton';
+import {
+  buildInspectorIssueContext,
+  buildLighthouseFailureContext,
+  buildRecommendationBulletContext,
+} from '@/lib/fixSuggestionContext';
 
 registerChartJsBase();
 
 export interface IssuesTabProps {
   lhData?: LinkLighthouseData | null;
   inspectorDetails: InspectorDetails | null;
+  pageUrl?: string;
 }
 
-export default function IssuesTab({ lhData, inspectorDetails }: IssuesTabProps) {
+export default function IssuesTab({ lhData, inspectorDetails, pageUrl }: IssuesTabProps) {
   const ci = strings.components.inspectorTabs;
   const it = strings.components.linkTabs.issues;
   const sj = strings.common;
@@ -160,8 +167,11 @@ export default function IssuesTab({ lhData, inspectorDetails }: IssuesTabProps)
                 <div className="text-xs text-muted-foreground mb-2">{it.lighthouseFailures}</div>
                 <div className="space-y-2">
                   {topFailures.map((f: LighthouseAuditRef, i: number) => (
-                    <div key={i} className="bg-brand-800 border border-default rounded-lg px-3 py-2 text-xs text-foreground">
-                      {f.helpText || f.id}
+                    <div key={i} className="bg-brand-800 border border-default rounded-lg px-3 py-2 text-xs text-foreground space-y-2">
+                      <span>{f.helpText || f.id}</span>
+                      <AiSuggestionButton
+                        request={buildLighthouseFailureContext(f.helpText || f.id || '', f.id, pageUrl)}
+                      />
                     </div>
                   ))}
                 </div>
@@ -240,6 +250,7 @@ export default function IssuesTab({ lhData, inspectorDetails }: IssuesTabProps)
                         {issue.recommendation}
                       </p>
                     ) : null}
+                    <AiSuggestionButton request={buildInspectorIssueContext(issue, pageUrl)} />
                   </div>
                 )}
               </div>
@@ -255,10 +266,13 @@ export default function IssuesTab({ lhData, inspectorDetails }: IssuesTabProps)
             {inspectorDetails.recommendations.map((rec: string, i: number) => (
               <div
                 key={i}
-                className="flex items-start gap-2 bg-brand-800 border border-default rounded-lg px-4 py-2.5"
+                className="flex flex-col gap-2 bg-brand-800 border border-default rounded-lg px-4 py-2.5"
               >
-                <ChevronRight className="h-3.5 w-3.5 text-link shrink-0 mt-0.5" />
-                <span className="text-sm text-foreground">{rec}</span>
+                <div className="flex items-start gap-2">
+                  <ChevronRight className="h-3.5 w-3.5 text-link shrink-0 mt-0.5" />
+                  <span className="text-sm text-foreground flex-1">{rec}</span>
+                </div>
+                <AiSuggestionButton request={buildRecommendationBulletContext(rec, pageUrl)} />
               </div>
             ))}
           </div>
diff --git a/web/src/components/links/tabs/PageAnalysisTab.tsx b/web/src/components/links/tabs/PageAnalysisTab.tsx
index da026c9f..95c076c6 100644
--- a/web/src/components/links/tabs/PageAnalysisTab.tsx
+++ b/web/src/components/links/tabs/PageAnalysisTab.tsx
@@ -9,6 +9,8 @@ import { linkHasBrowserErrors } from '@/lib/browserErrors';
 import { parseUrlTab } from '@/hooks/useUrlTab';
 import { severityBg } from '../../../utils/linkUtils';
 import { strings, format } from '../../../lib/strings';
+import AiSuggestionButton from '@/components/ai/AiSuggestionButton';
+import { buildOnPageWarningContext } from '@/lib/fixSuggestionContext';
 
 const SECTION_TABS = ['insights', 'browser', 'warnings', 'resources'] as const;
 type PageAnalysisSection = (typeof SECTION_TABS)[number];
@@ -251,7 +253,7 @@ function InsightsPanel({
   );
 }
 
-function WarningsPanel({ pa }: { pa: PageAnalysis }) {
+function WarningsPanel({ pa, pageUrl }: { pa: PageAnalysis; pageUrl: string }) {
   const p = strings.components.linkTabs.pageAnalysis;
   const [sevFilter, setSevFilter] = useState('All');
   const filteredWarnings = useMemo(() => {
@@ -282,15 +284,18 @@ function WarningsPanel({ pa }: { pa: PageAnalysis }) {
           {filteredWarnings.map((w, i) => (
             <li
               key={`${w.id}-${i}`}
-              className="bg-brand-900 border border-default rounded-lg px-3 py-2 text-sm text-foreground"
+              className="bg-brand-900 border border-default rounded-lg px-3 py-2 text-sm text-foreground space-y-2"
             >
-              <span className={`text-xs px-2 py-0.5 rounded mr-2 ${severityBg(w.severity)}`}>
-                {w.severity || 'info'}
-              </span>
-              {w.message}
-              {w.detail && (
-                <div className="mt-1 text-xs text-muted-foreground font-mono break-all">{w.detail}</div>
-              )}
+              <div>
+                <span className={`text-xs px-2 py-0.5 rounded mr-2 ${severityBg(w.severity)}`}>
+                  {w.severity || 'info'}
+                </span>
+                {w.message}
+                {w.detail && (
+                  <div className="mt-1 text-xs text-muted-foreground font-mono break-all">{w.detail}</div>
+                )}
+              </div>
+              <AiSuggestionButton request={buildOnPageWarningContext(w, pageUrl)} />
             </li>
           ))}
         </ul>
@@ -438,7 +443,7 @@ export default function PageAnalysisTab({ link }: PageAnalysisTabProps) {
         )}
         {activeSection === 'warnings' && (
           <ViewTabPanel idPrefix="page-analysis" tabId="warnings">
-            <WarningsPanel pa={pa} />
+            <WarningsPanel pa={pa} pageUrl={link.url} />
           </ViewTabPanel>
         )}
         {activeSection === 'resources' && (
diff --git a/web/src/components/links/tabs/SeoSocialTab.tsx b/web/src/components/links/tabs/SeoSocialTab.tsx
index 281290c3..2ebaad05 100644
--- a/web/src/components/links/tabs/SeoSocialTab.tsx
+++ b/web/src/components/links/tabs/SeoSocialTab.tsx
@@ -5,6 +5,7 @@ import { strings } from '../../../lib/strings';
 import { parseTechStack } from '../../../utils/linkUtils';
 import CopyBtn from '../CopyBtn';
 import OGPreview from '../OGPreview';
+import SerpPreview from '../SerpPreview';
 
 export interface SeoSocialTabProps {
   link: LinkDetail;
@@ -36,6 +37,10 @@ export default function SeoSocialTab({ link }: SeoSocialTabProps) {
 
   return (
     <div className="space-y-6">
+      <div>
+        <div className="text-xs text-muted-foreground mb-2">SERP snippet preview</div>
+        <SerpPreview link={link} />
+      </div>
       <div className="bg-brand-900 border border-default rounded-xl p-4 space-y-3">
         <div className="flex items-center justify-between">
           <div className="text-xs text-muted-foreground">{s.canonicalUrl}</div>
diff --git a/web/src/components/links/tabs/TechnicalTab.tsx b/web/src/components/links/tabs/TechnicalTab.tsx
index c2d3c028..d390c400 100644
--- a/web/src/components/links/tabs/TechnicalTab.tsx
+++ b/web/src/components/links/tabs/TechnicalTab.tsx
@@ -4,6 +4,8 @@ import type { LinkDetail } from '@/types/report';
 import { strings, format } from '../../../lib/strings';
 import SecHeaderRow from '../SecHeaderRow';
 import MiniBar from '../MiniBar';
+import AiSuggestionButton from '@/components/ai/AiSuggestionButton';
+import { buildTechnicalLinkIssueContext } from '@/lib/fixSuggestionContext';
 
 function headerPresent(val: unknown): boolean {
   if (val == null) return false;
@@ -62,6 +64,7 @@ export default function TechnicalTab({ link }: TechnicalTabProps) {
               label={h.label}
               value={String(link[h.field as keyof LinkDetail] ?? '')}
               recommendation={h.rec}
+              pageUrl={link.url}
             />
           ))}
         </div>
@@ -73,14 +76,22 @@ export default function TechnicalTab({ link }: TechnicalTabProps) {
         </h3>
         <div className="space-y-2">
           {perfRows.map(({ label, value, mono, warn }) => (
-            <div
-              key={label}
-              className="bg-brand-900 border border-default rounded-lg flex items-center justify-between px-4 py-2.5"
-            >
-              <span className="text-sm text-muted-foreground">{label}</span>
-              <span className={`text-sm ${mono ? 'font-mono' : ''} ${warn ? 'text-red-600 dark:text-red-400' : 'text-foreground'}`}>
-                {value}
-              </span>
+            <div key={label} className="bg-brand-900 border border-default rounded-lg px-4 py-2.5 space-y-2">
+              <div className="flex items-center justify-between">
+                <span className="text-sm text-muted-foreground">{label}</span>
+                <span className={`text-sm ${mono ? 'font-mono' : ''} ${warn ? 'text-red-600 dark:text-red-400' : 'text-foreground'}`}>
+                  {value}
+                </span>
+              </div>
+              {warn ? (
+                <AiSuggestionButton
+                  request={buildTechnicalLinkIssueContext(
+                    `${label}: ${value}`,
+                    link.url,
+                    'mixed_content',
+                  )}
+                />
+              ) : null}
             </div>
           ))}
         </div>
diff --git a/web/src/components/overview/OverviewHealthTab.tsx b/web/src/components/overview/OverviewHealthTab.tsx
index f8b37dd3..368259fc 100644
--- a/web/src/components/overview/OverviewHealthTab.tsx
+++ b/web/src/components/overview/OverviewHealthTab.tsx
@@ -1,5 +1,7 @@
 'use client';
 
+import Link from 'next/link';
+import { useSearchParams } from 'next/navigation';
 import { ChevronRight, Lightbulb } from 'lucide-react';
 import type { ReportCategory, ReportPayload } from '@/types';
 import { strings, format } from '@/lib/strings';
@@ -7,6 +9,9 @@ import { categoryDisplayName } from '@/lib/categoryDisplayNames';
 import { CategoryScoreGauge } from '@/components/charts/CategoryScoreGauge';
 import { Card } from '@/components';
 import { OverviewTabPanel } from './OverviewTabPanel';
+import { PortfolioBenchmarkCard } from './PortfolioBenchmarkCard';
+import AiSuggestionButton from '@/components/ai/AiSuggestionButton';
+import { buildOverviewRecommendationContext } from '@/lib/fixSuggestionContext';
 
 const REC_COLORS = [
   { border: 'border-l-blue-500', bg: 'bg-blue-500/10', text: 'text-link', dot: 'bg-blue-500' },
@@ -26,9 +31,21 @@ export interface OverviewHealthTabProps {
 export function OverviewHealthTab({ data, categoriesFiltered, recommendationsFiltered }: OverviewHealthTabProps) {
   const vo = strings.views.overview;
   const sj = strings.common;
+  const searchParams = useSearchParams();
+  const querySuffix = searchParams.toString() ? `?${searchParams.toString()}` : '';
+  const hasSubdomains = Boolean(data.subdomains && !data.subdomains.disabled);
+  const hasContacts = Boolean(
+    data.contact_intelligence &&
+      ((data.contact_intelligence.emails?.length ?? 0) +
+        (data.contact_intelligence.phones?.length ?? 0) +
+        (data.contact_intelligence.addresses?.length ?? 0) +
+        (data.contact_intelligence.organization_names?.length ?? 0) >
+        0),
+  );
 
   return (
     <OverviewTabPanel tabId="health" className="space-y-8">
+      <PortfolioBenchmarkCard benchmark={data.portfolio_benchmark} />
       <div>
         <h2 className="text-xl font-bold text-bright mb-4">{vo.healthByCategory}</h2>
         {data.categories && data.categories.length > 0 ? (
@@ -70,7 +87,50 @@ export function OverviewHealthTab({ data, categoriesFiltered, recommendationsFil
                     : ''}
               </span>
             </div>
+            {data.site_level.ads_txt_present != null && (
+              <div className="flex items-center gap-2">
+                <span className="text-muted-foreground">{vo.adsTxt}</span>
+                <span className="font-semibold text-foreground">
+                  {data.site_level.ads_txt_present ? sj.yes : sj.no}
+                  {data.site_level.ads_txt_present
+                    ? data.site_level.ads_txt_valid
+                      ? vo.adsTxtValid
+                      : vo.adsTxtInvalid
+                    : ''}
+                </span>
+              </div>
+            )}
+            {data.site_level.security_txt_present != null && (
+              <div className="flex items-center gap-2">
+                <span className="text-muted-foreground">{vo.securityTxt}</span>
+                <span className="font-semibold text-foreground">
+                  {data.site_level.security_txt_present ? sj.yes : sj.no}
+                  {data.site_level.security_txt_present &&
+                  (data.site_level.security_txt_contact?.length ?? 0) > 0
+                    ? format(vo.securityTxtContacts, {
+                        count: data.site_level.security_txt_contact?.length ?? 0,
+                      })
+                    : ''}
+                </span>
+              </div>
+            )}
           </Card>
+          {(hasSubdomains || hasContacts) && (
+            <div className="flex flex-wrap gap-4 mt-3 text-sm">
+              {hasSubdomains ? (
+                <Link href={`/subdomains${querySuffix}`} className="text-link hover:underline inline-flex items-center gap-1">
+                  {vo.viewSubdomains}
+                  <ChevronRight className="h-4 w-4" aria-hidden />
+                </Link>
+              ) : null}
+              {hasContacts ? (
+                <Link href={`/contacts${querySuffix}`} className="text-link hover:underline inline-flex items-center gap-1">
+                  {vo.viewContacts}
+                  <ChevronRight className="h-4 w-4" aria-hidden />
+                </Link>
+              ) : null}
+            </div>
+          )}
         </div>
       )}
 
@@ -86,10 +146,13 @@ export function OverviewHealthTab({ data, categoriesFiltered, recommendationsFil
                 return (
                   <div
                     key={i}
-                    className={`flex items-start gap-3 border-l-4 ${c.border} ${c.bg} rounded-r-xl px-4 py-3`}
+                    className={`flex flex-col gap-2 border-l-4 ${c.border} ${c.bg} rounded-r-xl px-4 py-3`}
                   >
-                    <ChevronRight className={`h-4 w-4 shrink-0 mt-0.5 ${c.text}`} />
-                    <span className="text-sm text-foreground leading-relaxed">{r}</span>
+                    <div className="flex items-start gap-3">
+                      <ChevronRight className={`h-4 w-4 shrink-0 mt-0.5 ${c.text}`} />
+                      <span className="text-sm text-foreground leading-relaxed flex-1">{r}</span>
+                    </div>
+                    <AiSuggestionButton request={buildOverviewRecommendationContext(r)} className="pl-7" />
                   </div>
                 );
               })}
diff --git a/web/src/components/overview/PortfolioBenchmarkCard.tsx b/web/src/components/overview/PortfolioBenchmarkCard.tsx
new file mode 100644
index 00000000..c598a789
--- /dev/null
+++ b/web/src/components/overview/PortfolioBenchmarkCard.tsx
@@ -0,0 +1,81 @@
+'use client';
+
+import { AlertCircle, TrendingUp } from 'lucide-react';
+import { Card } from '@/components';
+import { strings } from '@/lib/strings';
+import type { PortfolioBenchmark } from '@/types/report';
+
+interface PortfolioBenchmarkCardProps {
+  benchmark?: PortfolioBenchmark | null;
+}
+
+export function PortfolioBenchmarkCard({ benchmark }: PortfolioBenchmarkCardProps) {
+  const vo = strings.views.overview;
+  if (!benchmark) return null;
+
+  const status = benchmark.status;
+  const property = benchmark.property_health_score;
+  const median = benchmark.median_health_score;
+  const showScores = status === 'ok' || status == null;
+  const showBanner = status && status !== 'ok';
+
+  if (!showScores && !showBanner && property == null && median == null) return null;
+
+  const delta =
+    property != null && median != null ? property - median : null;
+
+  return (
+    <Card padding="tight" className="mb-8">
+      <div className="flex items-center gap-2 mb-3">
+        <TrendingUp className="h-4 w-4 text-link" />
+        <h2 className="text-xl font-bold text-bright">{vo.portfolioBenchmarkTitle}</h2>
+      </div>
+      <p className="text-xs text-muted-foreground mb-4">{vo.portfolioBenchmarkHint}</p>
+
+      {showBanner && benchmark.message ? (
+        <div
+          className={`mb-4 flex items-start gap-2 rounded-lg border px-3 py-2 text-sm ${
+            status === 'error'
+              ? 'border-red-500/30 bg-red-500/10 text-red-800 dark:text-red-300'
+              : 'border-amber-500/30 bg-amber-500/10 text-amber-800 dark:text-amber-200'
+          }`}
+        >
+          <AlertCircle className="h-4 w-4 shrink-0 mt-0.5" aria-hidden />
+          <span>{benchmark.message}</span>
+        </div>
+      ) : null}
+
+      {showScores ? (
+        <div className="grid grid-cols-1 sm:grid-cols-3 gap-4">
+          <div className="rounded-lg border border-default bg-brand-900/60 p-4">
+            <div className="text-xs uppercase tracking-wider text-muted-foreground">{vo.portfolioPropertyScore}</div>
+            <div className="text-3xl font-bold text-foreground mt-1">{property ?? '—'}</div>
+          </div>
+          <div className="rounded-lg border border-default bg-brand-900/60 p-4">
+            <div className="text-xs uppercase tracking-wider text-muted-foreground">{vo.portfolioMedianScore}</div>
+            <div className="text-3xl font-bold text-foreground mt-1">{median ?? '—'}</div>
+          </div>
+          <div className="rounded-lg border border-default bg-brand-900/60 p-4">
+            <div className="text-xs uppercase tracking-wider text-muted-foreground">{vo.portfolioDelta}</div>
+            <div
+              className={`text-3xl font-bold mt-1 ${
+                delta == null
+                  ? 'text-muted-foreground'
+                  : delta >= 0
+                    ? 'text-green-700 dark:text-green-400'
+                    : 'text-amber-700 dark:text-amber-400'
+              }`}
+            >
+              {delta == null ? '—' : `${delta >= 0 ? '+' : ''}${delta}`}
+            </div>
+          </div>
+        </div>
+      ) : property != null ? (
+        <div className="rounded-lg border border-default bg-brand-900/60 p-4 max-w-xs">
+          <div className="text-xs uppercase tracking-wider text-muted-foreground">{vo.portfolioPropertyScore}</div>
+          <div className="text-3xl font-bold text-foreground mt-1">{property}</div>
+        </div>
+      ) : null}
+    </Card>
+  );
+}
diff --git a/web/src/components/pipeline/PipelineLogViewer.tsx b/web/src/components/pipeline/PipelineLogViewer.tsx
index f3897934..afa0863b 100644
--- a/web/src/components/pipeline/PipelineLogViewer.tsx
+++ b/web/src/components/pipeline/PipelineLogViewer.tsx
@@ -46,7 +46,15 @@ function highlightText(text: string, query: string) {
   );
 }
 
-function ProgressLine({ line, query }: { line: PipelineLogLine; query: string }) {
+function ProgressLine({
+  line,
+  query,
+  label = 'Progress',
+}: {
+  line: PipelineLogLine;
+  query: string;
+  label?: string;
+}) {
   const p = line.progress;
   if (!p) {
     return (
@@ -56,7 +64,7 @@ function ProgressLine({ line, query }: { line: PipelineLogLine; query: string })
   return (
     <div className="space-y-1.5 rounded-md border border-blue-500/20 bg-blue-500/5 px-3 py-2">
       <div className="flex items-center justify-between gap-2 text-[11px] text-muted-foreground">
-        <span className="font-medium text-blue-200/90">Crawl progress</span>
+        <span className="font-medium text-blue-200/90">{label}</span>
         <span className="tabular-nums text-foreground/80">
           {p.current}/{p.total} ({p.percent}%)
         </span>
@@ -71,6 +79,22 @@ function ProgressLine({ line, query }: { line: PipelineLogLine; query: string })
   );
 }
 
+function ActivityLine({ line, query }: { line: PipelineLogLine; query: string }) {
+  const url = line.progressEvent?.url ?? line.text.replace(/^→\s*/, '');
+  return (
+    <div className="space-y-1.5 py-1">
+      <div
+        className="truncate rounded-md border border-default/60 bg-brand-900/50 px-2 py-1 font-mono text-[11px] text-[#e6edf3]/90"
+        title={url}
+      >
+        <span className="text-blue-300/90">→ </span>
+        {highlightText(url, query)}
+      </div>
+      {line.progress ? <ProgressLine line={line} query={query} label="Crawl progress" /> : null}
+    </div>
+  );
+}
+
 function LogLine({ line, query }: { line: PipelineLogLine; query: string }) {
   if (line.kind === 'noise') {
     return (
@@ -80,10 +104,18 @@ function LogLine({ line, query }: { line: PipelineLogLine; query: string }) {
     );
   }
 
+  if (line.kind === 'activity') {
+    return <ActivityLine line={line} query={query} />;
+  }
+
   if (line.kind === 'progress') {
     return (
       <div className="py-1">
-        <ProgressLine line={line} query={query} />
+        <ProgressLine
+          line={line}
+          query={query}
+          label={line.phase === 'crawl' ? 'Crawl progress' : 'Progress'}
+        />
       </div>
     );
   }
diff --git a/web/src/components/pipeline/PipelineProgressHeader.tsx b/web/src/components/pipeline/PipelineProgressHeader.tsx
new file mode 100644
index 00000000..bb2931f9
--- /dev/null
+++ b/web/src/components/pipeline/PipelineProgressHeader.tsx
@@ -0,0 +1,162 @@
+'use client';
+
+import { useMemo } from 'react';
+import { Check, Loader2 } from 'lucide-react';
+import type { PipelineJobStatus } from '@/types/api';
+import type { LivePipelineEstimate } from '@/lib/pipelineLiveEstimate';
+import {
+  PHASE_LABELS,
+  PIPELINE_STEPPER_PHASES,
+  computeEta,
+  formatDurationMs,
+  parsePipelineProgressEvents,
+  resolveActiveProgress,
+  stepLabel,
+  type ProgressPhase,
+} from '@/lib/formatPipelineLog';
+
+export interface PipelineProgressHeaderProps {
+  log: string;
+  status?: PipelineJobStatus | '';
+  liveEstimate?: LivePipelineEstimate | null;
+  compact?: boolean;
+  className?: string;
+}
+
+function truncateUrl(url: string, max = 56): string {
+  if (url.length <= max) return url;
+  return `${url.slice(0, max - 1)}…`;
+}
+
+function phaseIndex(phase: ProgressPhase): number {
+  const idx = PIPELINE_STEPPER_PHASES.indexOf(phase);
+  return idx >= 0 ? idx : -1;
+}
+
+export default function PipelineProgressHeader({
+  log,
+  status = '',
+  liveEstimate = null,
+  compact = false,
+  className = '',
+}: PipelineProgressHeaderProps) {
+  const events = useMemo(() => parsePipelineProgressEvents(log), [log]);
+  const latest = useMemo(() => resolveActiveProgress(events, status), [events, status]);
+  const eta = useMemo(() => computeEta(latest, events), [latest, events]);
+
+  if (!latest) return null;
+
+  const jobFinished = status === 'success' || status === 'error';
+  const activePhase = latest.phase;
+  const activeIdx = jobFinished && latest.step === 'done' ? PIPELINE_STEPPER_PHASES.length : phaseIndex(activePhase);
+  const stepText = stepLabel(latest.step, latest.message);
+  const phaseLabel = PHASE_LABELS[activePhase] ?? activePhase;
+  const isActive = !jobFinished && latest.step !== 'done';
+  const hasBar =
+    isActive &&
+    latest.current != null &&
+    latest.total != null &&
+    latest.total > 0;
+  const barPct = hasBar
+    ? Math.min(100, Math.round(((latest.current ?? 0) / (latest.total ?? 1)) * 100))
+    : null;
+
+  return (
+    <div
+      className={`rounded-lg border border-default bg-brand-900/80 ${compact ? 'px-2 py-2' : 'px-3 py-3'} ${className}`}
+      role="status"
+      aria-live="polite"
+    >
+      {!compact ? (
+        <div className="mb-3 flex flex-wrap items-center gap-1.5">
+          {PIPELINE_STEPPER_PHASES.map((phase, i) => {
+            const done = jobFinished && latest.step === 'done' ? true : activeIdx >= 0 && i < activeIdx;
+            const active = !jobFinished && phase === activePhase && latest.step !== 'done';
+            const future = !done && !active;
+            return (
+              <div key={phase} className="flex items-center gap-1.5">
+                <span
+                  className={`rounded-full px-2 py-0.5 text-[10px] font-semibold uppercase tracking-wide ${
+                    active
+                      ? 'bg-blue-500/20 text-blue-200 ring-1 ring-blue-500/40'
+                      : done
+                        ? 'bg-green-500/15 text-green-300'
+                        : future
+                          ? 'text-muted-foreground/60'
+                          : 'text-muted-foreground'
+                  }`}
+                >
+                  {PHASE_LABELS[phase]}
+                </span>
+                {i < PIPELINE_STEPPER_PHASES.length - 1 ? (
+                  <span className="text-muted-foreground/40 text-[10px]">›</span>
+                ) : null}
+              </div>
+            );
+          })}
+        </div>
+      ) : null}
+
+      <div className={`flex flex-wrap items-center justify-between gap-2 ${compact ? 'text-[10px]' : 'text-xs'}`}>
+        <div className="flex min-w-0 items-center gap-2">
+          {isActive ? (
+            <Loader2 className="h-3.5 w-3.5 shrink-0 animate-spin text-link" aria-hidden />
+          ) : jobFinished && status === 'success' ? (
+            <Check className="h-3.5 w-3.5 shrink-0 text-green-500" aria-hidden />
+          ) : null}
+          <span className="font-medium text-foreground">
+            {phaseLabel}
+            {!compact ? ` · ${stepText}` : `: ${stepText}`}
+          </span>
+        </div>
+        <div className="flex shrink-0 flex-wrap items-center gap-2 tabular-nums text-muted-foreground">
+          {hasBar ? (
+            <span>
+              {latest.current}/{latest.total}
+              {barPct != null ? ` (${barPct}%)` : ''}
+            </span>
+          ) : null}
+          {eta.ratePerSec != null && latest.phase === 'crawl' ? (
+            <span>{eta.ratePerSec.toFixed(1)} pg/s</span>
+          ) : null}
+          {eta.elapsedMs != null ? <span>elapsed {formatDurationMs(eta.elapsedMs)}</span> : null}
+          {eta.remainingMs != null && !liveEstimate?.remainingMs ? (
+            <span className="text-foreground/80">step ETA {formatDurationMs(eta.remainingMs)}</span>
+          ) : null}
+          {liveEstimate?.remainingMs != null && isActive ? (
+            <span className="font-medium text-foreground/90">
+              ~{formatDurationMs(liveEstimate.remainingMs)} left total
+            </span>
+          ) : null}
+        </div>
+      </div>
+
+      {liveEstimate?.ratePerSec != null && isActive && latest.phase === 'crawl' && !compact ? (
+        <p className="mt-1.5 text-[10px] text-muted-foreground">
+          Avg {liveEstimate.ratePerSec.toFixed(2)} pages/s
+          {liveEstimate.observedCrawlPages != null ? ` · ${liveEstimate.observedCrawlPages} crawled` : ''}
+          {liveEstimate.totalMs != null ? ` · ~${formatDurationMs(liveEstimate.totalMs)} projected total` : ''}
+        </p>
+      ) : null}
+
+      {hasBar && barPct != null ? (
+        <div className={`${compact ? 'mt-1.5' : 'mt-2'} h-1.5 overflow-hidden rounded-full bg-brand-700/80`}>
+          <div
+            className="h-full rounded-full bg-gradient-to-r from-blue-600 to-sky-400 transition-[width] duration-300 ease-out"
+            style={{ width: `${barPct}%` }}
+          />
+        </div>
+      ) : latest.step !== 'done' ? (
+        <div className={`${compact ? 'mt-1.5' : 'mt-2'} h-1.5 overflow-hidden rounded-full bg-brand-700/80`}>
+          <div className="h-full w-1/3 animate-pulse rounded-full bg-blue-500/50" />
+        </div>
+      ) : null}
+
+      {latest.url && !compact ? (
+        <p className="mt-2 truncate font-mono text-[11px] text-muted-foreground" title={latest.url}>
+          {truncateUrl(latest.url)}
+        </p>
+      ) : null}
+    </div>
+  );
+}
diff --git a/web/src/components/pipeline/PipelineRunPanel.tsx b/web/src/components/pipeline/PipelineRunPanel.tsx
index 8e228b29..dd622378 100644
--- a/web/src/components/pipeline/PipelineRunPanel.tsx
+++ b/web/src/components/pipeline/PipelineRunPanel.tsx
@@ -1,6 +1,6 @@
 'use client';
 
-import { useEffect, useRef, useState, type KeyboardEvent } from 'react';
+import { useEffect, useMemo, useRef, useState, type KeyboardEvent } from 'react';
 import {
   ArrowLeft,
   ArrowRight,
@@ -27,8 +27,12 @@ import { PIPELINE_PRESETS } from './pipelinePresets';
 import { CRAWL_PRESETS, type CrawlPresetId } from '@/lib/crawlPresets';
 import PipelineWizardProgress, { type WizardStep } from './PipelineWizardProgress';
 import PipelineLogViewer from './PipelineLogViewer';
+import PipelineProgressHeader from './PipelineProgressHeader';
 import CrawlAuthorizeCheckbox from './CrawlAuthorizeCheckbox';
 import PipelineRunPreviewCard from './PipelineRunPreviewCard';
+import { buildPipelineRunPreview } from '@/lib/pipelineRunPreview';
+import { computeLivePipelineEstimate } from '@/lib/pipelineLiveEstimate';
+import { parsePipelineProgressEvents, resolveActiveProgress } from '@/lib/formatPipelineLog';
 
 const s = strings.pipelineRunner;
 const crawlPresets = s.crawlPresets as Record<string, { label: string; maxPages: string; stream?: boolean }>;
@@ -108,6 +112,25 @@ export default function PipelineRunPanel() {
     presetId === 'crawl-only' ? strings.reportSelector.crawlOnlyNote : null;
   const showProgress = busy || Boolean(status) || Boolean(log);
 
+  const runPreview = useMemo(
+    () =>
+      buildPipelineRunPreview({
+        presetId,
+        configState,
+        customCommand,
+        crawlPresetId,
+      }),
+    [presetId, configState, customCommand, crawlPresetId],
+  );
+
+  const liveEstimate = useMemo(() => {
+    const isRunning = busy || status === 'running' || status === 'starting';
+    if (!isRunning || !log.trim()) return null;
+    const events = parsePipelineProgressEvents(log);
+    const latest = resolveActiveProgress(events, status);
+    return computeLivePipelineEstimate(runPreview, events, latest, status);
+  }, [busy, status, log, runPreview]);
+
   const goToStep = (next: WizardStep) => {
     setStep(next);
     setMaxStep((prev) => (next > prev ? next : prev));
@@ -329,6 +352,7 @@ export default function PipelineRunPanel() {
               configState={configState}
               customCommand={customCommand}
               crawlPresetId={crawlPresetId}
+              liveEstimate={liveEstimate}
             />
 
             <div className="mt-5">
@@ -408,6 +432,15 @@ export default function PipelineRunPanel() {
                 </div>
               </div>
 
+              {log ? (
+                <PipelineProgressHeader
+                  log={log}
+                  status={status}
+                  liveEstimate={liveEstimate}
+                  className="mb-3"
+                />
+              ) : null}
+
               {log ? (
                 <div>
                   <button
diff --git a/web/src/components/pipeline/PipelineRunPreviewCard.tsx b/web/src/components/pipeline/PipelineRunPreviewCard.tsx
index 75f19014..570faa19 100644
--- a/web/src/components/pipeline/PipelineRunPreviewCard.tsx
+++ b/web/src/components/pipeline/PipelineRunPreviewCard.tsx
@@ -7,6 +7,7 @@ import {
   buildPipelineRunPreview,
   formatPipelineRunDuration,
 } from '@/lib/pipelineRunPreview';
+import { formatLivePipelineDuration, type LivePipelineEstimate } from '@/lib/pipelineLiveEstimate';
 import type { PipelinePresetId } from '@/components/pipeline/pipelinePresets';
 import type { CrawlPresetId } from '@/lib/crawlPresets';
 import type { PipelineConfigState } from '@/types/api';
@@ -18,6 +19,7 @@ export interface PipelineRunPreviewCardProps {
   configState: PipelineConfigState;
   customCommand?: string;
   crawlPresetId?: CrawlPresetId | '';
+  liveEstimate?: LivePipelineEstimate | null;
 }
 
 export default function PipelineRunPreviewCard({
@@ -25,6 +27,7 @@ export default function PipelineRunPreviewCard({
   configState,
   customCommand = '',
   crawlPresetId = '',
+  liveEstimate = null,
 }: PipelineRunPreviewCardProps) {
   const [configOpen, setConfigOpen] = useState(false);
   const preview = buildPipelineRunPreview({
@@ -34,7 +37,10 @@ export default function PipelineRunPreviewCard({
     crawlPresetId,
   });
 
-  const duration = formatPipelineRunDuration(preview.timeMinSeconds, preview.timeMaxSeconds);
+  const isLive = liveEstimate != null && liveEstimate.source === 'live' && (liveEstimate.remainingMs ?? 0) > 0;
+  const duration = isLive
+    ? formatLivePipelineDuration(liveEstimate)
+    : formatPipelineRunDuration(preview.timeMinSeconds, preview.timeMaxSeconds);
 
   return (
     <div className="mt-5 space-y-4 rounded-xl border border-default bg-brand-900/40 p-4 sm:p-5">
@@ -43,9 +49,17 @@ export default function PipelineRunPreviewCard({
           <h3 className="text-sm font-semibold text-foreground">{s.title}</h3>
           <p className="mt-1 text-xs leading-relaxed text-muted-foreground">{s.hint}</p>
         </div>
-        <div className="inline-flex items-center gap-1.5 rounded-full border border-blue-500/30 bg-blue-500/10 px-3 py-1 text-xs font-medium text-blue-800 dark:text-blue-200">
+        <div
+          className={`inline-flex max-w-full items-center gap-1.5 rounded-full border px-3 py-1 text-xs font-medium ${
+            isLive
+              ? 'border-green-500/35 bg-green-500/10 text-green-800 dark:text-green-200'
+              : 'border-blue-500/30 bg-blue-500/10 text-blue-800 dark:text-blue-200'
+          }`}
+        >
           <Clock className="h-3.5 w-3.5 shrink-0" aria-hidden />
-          {s.estimatedTime}: {duration}
+          <span className="truncate">
+            {isLive ? s.liveEstimatedTime : s.estimatedTime}: {duration}
+          </span>
         </div>
       </div>
 
diff --git a/web/src/components/pipeline/PipelineRunnerFab.tsx b/web/src/components/pipeline/PipelineRunnerFab.tsx
index 1f8e47ba..b9527c7e 100644
--- a/web/src/components/pipeline/PipelineRunnerFab.tsx
+++ b/web/src/components/pipeline/PipelineRunnerFab.tsx
@@ -6,6 +6,7 @@ import { strings } from '@/lib/strings';
 import { usePipeline } from '@/context/PipelineContext';
 import { useSession } from '@/context/SessionContext';
 import { storePipelineReturnPath } from '@/lib/pipelineReturn';
+import PipelineProgressHeader from './PipelineProgressHeader';
 
 const s = strings.pipelineRunner;
 
@@ -39,11 +40,13 @@ export default function PipelineRunnerFab() {
   return (
     <div className="print:hidden fixed bottom-6 right-6 z-50 flex flex-col items-end gap-2">
       {showDock ? (
-        <div
-          role="status"
-          aria-live="polite"
-          className="flex max-w-[min(100vw-2rem,20rem)] items-center gap-3 rounded-xl border border-default bg-brand-800 px-3 py-2.5 shadow-xl"
-        >
+        <div className="flex max-w-[min(100vw-2rem,22rem)] flex-col gap-2">
+          {log && busy ? <PipelineProgressHeader log={log} compact className="shadow-xl" /> : null}
+          <div
+            role="status"
+            aria-live="polite"
+            className="flex items-center gap-3 rounded-xl border border-default bg-brand-800 px-3 py-2.5 shadow-xl"
+          >
           {busy ? (
             <Loader2 className="h-5 w-5 shrink-0 animate-spin text-link" aria-hidden />
           ) : (
@@ -89,6 +92,7 @@ export default function PipelineRunnerFab() {
           >
             <Maximize2 className="h-4 w-4" />
           </button>
+          </div>
         </div>
       ) : null}
       <button
diff --git a/web/src/components/siteStructure/CrawlMapPanel.tsx b/web/src/components/siteStructure/CrawlMapPanel.tsx
new file mode 100644
index 00000000..1775c226
--- /dev/null
+++ b/web/src/components/siteStructure/CrawlMapPanel.tsx
@@ -0,0 +1,186 @@
+'use client';
+
+import { useCallback, useMemo, useState } from 'react';
+import { ChevronDown, ChevronRight } from 'lucide-react';
+import type { CrawlSegmentsData, PathTreeNode } from '@/types/report';
+
+interface CrawlMapPanelProps {
+  tree: PathTreeNode;
+  crawlSegments?: CrawlSegmentsData | null;
+  selectedPath?: string | null;
+  onSelectPath?: (pathKey: string) => void;
+  maxNodes?: number;
+}
+
+function healthBarClass(score: number | null | undefined): string {
+  if (score == null || !Number.isFinite(score)) return 'bg-cyan-500/40';
+  if (score < 50) return 'bg-red-500/50';
+  if (score < 70) return 'bg-amber-500/50';
+  return 'bg-emerald-500/50';
+}
+
+function barWidth(pages: number, maxPages: number): number {
+  if (maxPages <= 0) return 8;
+  return Math.max(8, Math.round((pages / maxPages) * 120));
+}
+
+function TreeNodeRow({
+  node,
+  depth,
+  expanded,
+  selectedPath,
+  maxPages,
+  onToggle,
+  onSelect,
+  visibleCount,
+  maxNodes,
+  healthByPrefix,
+}: {
+  node: PathTreeNode;
+  depth: number;
+  expanded: Set<string>;
+  selectedPath?: string | null;
+  maxPages: number;
+  onToggle: (key: string) => void;
+  onSelect: (key: string) => void;
+  visibleCount: { n: number };
+  maxNodes: number;
+  healthByPrefix: Map<string, number | null | undefined>;
+}) {
+  if (visibleCount.n >= maxNodes) return null;
+  visibleCount.n += 1;
+
+  const key = node.pathKey || '/';
+  const hasChildren = (node.children?.length ?? 0) > 0;
+  const isOpen = expanded.has(key);
+  const pages = node.current?.pages ?? 0;
+  const isSelected = selectedPath === key;
+
+  return (
+    <>
+      <button
+        type="button"
+        onClick={() => onSelect(key)}
+        className={`flex w-full items-center gap-2 rounded-md px-2 py-1.5 text-left text-xs transition-colors hover:bg-brand-800/80 ${
+          isSelected ? 'bg-blue-500/15 ring-1 ring-blue-500/30' : ''
+        }`}
+        style={{ paddingLeft: `${8 + depth * 16}px` }}
+      >
+        {hasChildren ? (
+          <span
+            role="presentation"
+            className="shrink-0 text-muted-foreground"
+            onClick={(e) => {
+              e.stopPropagation();
+              onToggle(key);
+            }}
+          >
+            {isOpen ? <ChevronDown className="h-3.5 w-3.5" /> : <ChevronRight className="h-3.5 w-3.5" />}
+          </span>
+        ) : (
+          <span className="w-3.5 shrink-0" />
+        )}
+        <span className="font-mono text-foreground truncate flex-1">{node.segment || key}</span>
+        <span className="tabular-nums text-muted-foreground shrink-0">{pages}</span>
+        <span
+          className={`h-2 rounded-sm shrink-0 ${healthBarClass(healthByPrefix.get(key))}`}
+          style={{ width: `${barWidth(pages, maxPages)}px` }}
+          aria-hidden
+        />
+      </button>
+      {hasChildren && isOpen
+        ? node.children!.map((child) => (
+            <TreeNodeRow
+              key={child.pathKey}
+              node={child}
+              depth={depth + 1}
+              expanded={expanded}
+              selectedPath={selectedPath}
+              maxPages={maxPages}
+              onToggle={onToggle}
+              onSelect={onSelect}
+              visibleCount={visibleCount}
+              maxNodes={maxNodes}
+              healthByPrefix={healthByPrefix}
+            />
+          ))
+        : null}
+    </>
+  );
+}
+
+function collectMaxPages(node: PathTreeNode): number {
+  let max = node.current?.pages ?? 0;
+  for (const child of node.children || []) {
+    max = Math.max(max, collectMaxPages(child));
+  }
+  return max;
+}
+
+export default function CrawlMapPanel({
+  tree,
+  crawlSegments = null,
+  selectedPath = null,
+  onSelectPath,
+  maxNodes = 80,
+}: CrawlMapPanelProps) {
+  const [expanded, setExpanded] = useState<Set<string>>(() => new Set(['/']));
+
+  const maxPages = useMemo(() => collectMaxPages(tree), [tree]);
+  const healthByPrefix = useMemo(() => {
+    const m = new Map<string, number | null | undefined>();
+    for (const seg of crawlSegments?.segments || []) {
+      const prefix = seg.prefix?.startsWith('/') ? seg.prefix : `/${seg.prefix || ''}`;
+      m.set(prefix === '//' ? '/' : prefix.replace(/\/+$/, '') || '/', seg.health_score);
+    }
+    return m;
+  }, [crawlSegments]);
+
+  const toggle = useCallback((key: string) => {
+    setExpanded((prev) => {
+      const next = new Set(prev);
+      if (next.has(key)) next.delete(key);
+      else next.add(key);
+      return next;
+    });
+  }, []);
+
+  const select = useCallback(
+    (key: string) => {
+      onSelectPath?.(key);
+      if (key !== '/') {
+        setExpanded((prev) => {
+          const next = new Set(prev);
+          next.add('/');
+          next.add(key);
+          return next;
+        });
+      }
+    },
+    [onSelectPath],
+  );
+
+  const counter = { n: 0 };
+
+  return (
+    <div className="rounded-xl border border-default bg-brand-900/40 p-4">
+      <p className="text-xs text-muted-foreground mb-3">
+        Click a path segment to filter the structure tree. Expand nodes to explore depth.
+      </p>
+      <div className="max-h-[min(520px,60vh)] overflow-y-auto">
+        <TreeNodeRow
+          node={tree}
+          depth={0}
+          expanded={expanded}
+          selectedPath={selectedPath}
+          maxPages={maxPages}
+          onToggle={toggle}
+          onSelect={select}
+          visibleCount={counter}
+          maxNodes={maxNodes}
+          healthByPrefix={healthByPrefix}
+        />
+      </div>
+    </div>
+  );
+}
diff --git a/web/src/components/siteStructure/SiteStructureLinkGraph.tsx b/web/src/components/siteStructure/SiteStructureLinkGraph.tsx
new file mode 100644
index 00000000..bf064512
--- /dev/null
+++ b/web/src/components/siteStructure/SiteStructureLinkGraph.tsx
@@ -0,0 +1,142 @@
+'use client';
+
+import { useEffect, useMemo, useRef } from 'react';
+import Link from 'next/link';
+import ForceGraph3D from '3d-force-graph';
+import { useReport } from '@/context/useReport';
+import type { GraphEdge, GraphNode } from '@/types';
+
+const MAX_NODES = 200;
+
+interface GraphNodeData {
+  id: string;
+  label: string;
+  title: string;
+  color: string;
+}
+
+interface GraphLinkData {
+  source: string;
+  target: string;
+}
+
+interface ForceGraphInstance {
+  _destructor?: () => void;
+  width: (w: number) => ForceGraphInstance;
+  height: (h: number) => ForceGraphInstance;
+  graphData: (data: { nodes: GraphNodeData[]; links: GraphLinkData[] }) => ForceGraphInstance;
+  nodeColor: (fn: (node: GraphNodeData) => string) => ForceGraphInstance;
+  nodeLabel: (fn: (node: GraphNodeData) => string) => ForceGraphInstance;
+  linkColor: (fn: () => string) => ForceGraphInstance;
+  onNodeClick: (fn: (node: GraphNodeData) => void) => ForceGraphInstance;
+  backgroundColor: (color: string) => ForceGraphInstance;
+}
+
+export default function SiteStructureLinkGraph() {
+  const { data } = useReport();
+  const containerRef = useRef<HTMLDivElement | null>(null);
+  const graphRef = useRef<ForceGraphInstance | null>(null);
+
+  const graphData = useMemo(() => {
+    if (!data) return null;
+    const urlToStatus: Record<string, string> = {};
+    (data.links || []).forEach((l) => {
+      urlToStatus[l.url] = String(l.status);
+    });
+    const nodes = (data.graph_nodes || []) as GraphNode[];
+    const edges = (data.graph_edges || []) as GraphEdge[];
+    if (!nodes.length && !edges.length) return null;
+
+    const nodeMap = new Map<string, GraphNodeData>();
+    nodes.slice(0, MAX_NODES).forEach((u) => {
+      const id = typeof u === 'string' ? u : (u.id || u.url || String(u));
+      const st = urlToStatus[id] || '';
+      const color = /^[45]/.test(st) ? '#EF4444' : /^2/.test(st) ? '#3B82F6' : '#64748b';
+      const label = String(id).replace(/^https?:\/\/[^/]+/, '') || '/';
+      nodeMap.set(id, { id, label, title: id, color });
+    });
+    const idSet = new Set(nodeMap.keys());
+    const links = edges
+      .map((e) => {
+        const fromId = e.from ?? e['from'];
+        const toId = e.to ?? e['to'];
+        return fromId && toId && idSet.has(fromId) && idSet.has(toId)
+          ? { source: fromId, target: toId }
+          : null;
+      })
+      .filter((l): l is GraphLinkData => l != null);
+
+    return {
+      nodes: [...nodeMap.values()],
+      links,
+      total: nodes.length,
+    };
+  }, [data]);
+
+  useEffect(() => {
+    const prev = graphRef.current;
+    if (prev?._destructor) {
+      try {
+        prev._destructor();
+      } catch {
+        /* ignore */
+      }
+    }
+    graphRef.current = null;
+
+    if (!containerRef.current || !graphData) return;
+
+    const el = containerRef.current;
+    type ForceGraphFactory = () => (container: HTMLElement) => ForceGraphInstance & {
+      graphData: (data: { nodes: GraphNodeData[]; links: GraphLinkData[] }) => ForceGraphInstance & Record<string, unknown>;
+      nodeColor: (fn: (node: GraphNodeData) => string) => ForceGraphInstance & Record<string, unknown>;
+      nodeLabel: (fn: (node: GraphNodeData) => string) => ForceGraphInstance & Record<string, unknown>;
+      linkColor: (fn: () => string) => ForceGraphInstance & Record<string, unknown>;
+      onNodeClick: (fn: (node: GraphNodeData) => void) => ForceGraphInstance & Record<string, unknown>;
+      backgroundColor: (color: string) => ForceGraphInstance & Record<string, unknown>;
+      width: (w: number) => ForceGraphInstance & Record<string, unknown>;
+      height: (h: number) => ForceGraphInstance & Record<string, unknown>;
+      _destructor?: () => void;
+    };
+    const createGraph = ForceGraph3D as unknown as ForceGraphFactory;
+    const graph = createGraph()(el)
+      .width(el.clientWidth)
+      .height(360)
+      .graphData(graphData)
+      .nodeColor((node: GraphNodeData) => node.color)
+      .nodeLabel((node: GraphNodeData) => node.label)
+      .linkColor(() => 'rgba(148,163,184,0.35)')
+      .backgroundColor('rgba(15,23,42,0.2)')
+      .onNodeClick((node: GraphNodeData) => {
+        if (node?.id) window.open(node.id, '_blank');
+      });
+
+    graphRef.current = graph;
+
+    return () => {
+      if (graph?._destructor) {
+        try {
+          graph._destructor();
+        } catch {
+          /* ignore */
+        }
+      }
+    };
+  }, [graphData]);
+
+  if (!graphData) {
+    return <p className="text-sm text-muted-foreground py-8 text-center">No link graph data in this report.</p>;
+  }
+
+  return (
+    <div className="space-y-3">
+      <p className="text-xs text-muted-foreground">
+        Showing up to {MAX_NODES} of {graphData.total} nodes. Click a node to open the URL.{' '}
+        <Link href="/network" className="text-link hover:underline">
+          Open full Network view
+        </Link>
+      </p>
+      <div ref={containerRef} className="h-[360px] w-full rounded-xl border border-default overflow-hidden" />
+    </div>
+  );
+}
diff --git a/web/src/lib/appNav.ts b/web/src/lib/appNav.ts
index d4144204..9c093bf5 100644
--- a/web/src/lib/appNav.ts
+++ b/web/src/lib/appNav.ts
@@ -22,6 +22,8 @@ import {
   TrendingUp,
   FileSearch,
   MessageSquare,
+  Globe2,
+  Contact2,
 } from 'lucide-react';
 import { strings } from '@/lib/strings';
 import { viewIdToPathSlug, type ViewId } from '@/routes';
@@ -56,6 +58,8 @@ const VIEW_NAV: { id: ViewId; icon: LucideIcon }[] = [
   { id: 'gallery', icon: Images },
   { id: 'search-performance', icon: TrendingUp },
   { id: 'indexation', icon: FileSearch },
+  { id: 'subdomains', icon: Globe2 },
+  { id: 'contacts', icon: Contact2 },
   { id: 'backlinks', icon: Link2 },
   { id: 'traffic', icon: BarChart2 },
   { id: 'keywords-explorer', icon: Key },
diff --git a/web/src/lib/customFields.test.ts b/web/src/lib/customFields.test.ts
new file mode 100644
index 00000000..ec8ba4ee
--- /dev/null
+++ b/web/src/lib/customFields.test.ts
@@ -0,0 +1,18 @@
+import { describe, expect, it } from 'vitest';
+import { collectCustomFieldKeys, parseLinkCustomFields } from '@/lib/customFields';
+import type { ReportLink } from '@/types';
+
+describe('customFields', () => {
+  it('parses JSON custom_fields string', () => {
+    const link = { url: 'https://ex.com', custom_fields: '{"price":"9.99","sku":"A1"}' } as ReportLink;
+    expect(parseLinkCustomFields(link)).toEqual({ price: '9.99', sku: 'A1' });
+  });
+
+  it('collects sorted keys across links', () => {
+    const links = [
+      { url: 'https://ex.com/a', custom_fields: '{"b":"2","a":"1"}' } as ReportLink,
+      { url: 'https://ex.com/b', custom_fields: '{"c":"3"}' } as ReportLink,
+    ];
+    expect(collectCustomFieldKeys(links)).toEqual(['a', 'b', 'c']);
+  });
+});
diff --git a/web/src/lib/customFields.ts b/web/src/lib/customFields.ts
new file mode 100644
index 00000000..15e0c7b0
--- /dev/null
+++ b/web/src/lib/customFields.ts
@@ -0,0 +1,28 @@
+import type { ReportLink } from '@/types';
+
+export function parseLinkCustomFields(link: ReportLink): Record<string, string> {
+  const raw = link.custom_fields;
+  if (!raw) return {};
+  if (typeof raw === 'object') {
+    return Object.fromEntries(Object.entries(raw).map(([k, v]) => [k, String(v ?? '')]));
+  }
+  const text = String(raw).trim();
+  if (!text) return {};
+  try {
+    const parsed = JSON.parse(text) as unknown;
+    if (!parsed || typeof parsed !== 'object' || Array.isArray(parsed)) return {};
+    return Object.fromEntries(
+      Object.entries(parsed as Record<string, unknown>).map(([k, v]) => [k, String(v ?? '')]),
+    );
+  } catch {
+    return {};
+  }
+}
+
+export function collectCustomFieldKeys(links: ReportLink[]): string[] {
+  const keys = new Set<string>();
+  for (const link of links) {
+    Object.keys(parseLinkCustomFields(link)).forEach((k) => keys.add(k));
+  }
+  return [...keys].sort();
+}
diff --git a/web/src/lib/exportAudit.ts b/web/src/lib/exportAudit.ts
index 05b00fcc..afb1b345 100644
--- a/web/src/lib/exportAudit.ts
+++ b/web/src/lib/exportAudit.ts
@@ -12,3 +12,17 @@ export function buildAuditExportUrl(
   if (options?.inline) q.set('disposition', 'inline');
   return reportApi(`/export?${q.toString()}`);
 }
+
+export function buildWorkbookExportUrl(reportId: number | null | undefined): string {
+  const q = new URLSearchParams();
+  if (reportId != null) q.set('reportId', String(reportId));
+  const qs = q.toString();
+  return reportApi(`/export-workbook${qs ? `?${qs}` : ''}`);
+}
+
+export function buildSitemapExportUrl(reportId: number | null | undefined): string {
+  const q = new URLSearchParams();
+  if (reportId != null) q.set('reportId', String(reportId));
+  const qs = q.toString();
+  return reportApi(`/export-sitemap${qs ? `?${qs}` : ''}`);
+}
diff --git a/web/src/lib/fixSuggestionContext.ts b/web/src/lib/fixSuggestionContext.ts
new file mode 100644
index 00000000..7df84374
--- /dev/null
+++ b/web/src/lib/fixSuggestionContext.ts
@@ -0,0 +1,334 @@
+import type { FixSuggestionRequest } from '@/types/fixSuggestion';
+import type { ReportIssue, SecurityFinding, LighthouseDiagnostic, LighthouseQuickWin, LighthouseAuditRef, RichResultsValidationRow, ReportRedirect } from '@/types/report';
+import type { QueryPageMisalignmentItem, CannibalisationItem } from '@/types/components';
+import type { CompetitorKeywordGapRow } from '@/types/report';
+import type { FlatBrowserErrorRow } from '@/lib/browserErrors';
+import type { IssueDeltaRow } from '@/lib/reportCompareExtras';
+import type { InspectorIssueRow } from '@/types/report';
+import type { ContentDuplicateCluster } from '@/types/report';
+
+export function buildIssueContext(issue: ReportIssue, category: string): FixSuggestionRequest {
+  return {
+    source: 'issue',
+    message: issue.message || 'Audit issue',
+    url: issue.url,
+    context: {
+      priority: issue.priority,
+      category,
+      recommendation: issue.recommendation,
+      type: issue.type || issue.finding_type,
+    },
+  };
+}
+
+export function buildLighthouseQuickWinContext(win: LighthouseQuickWin, passed: boolean): FixSuggestionRequest {
+  return {
+    source: 'lighthouse',
+    message: win.title || 'Lighthouse quick win',
+    context: {
+      kind: 'quick_win',
+      why: win.why,
+      how: win.how,
+      impact: win.impact,
+      passed,
+    },
+  };
+}
+
+export function buildLighthouseDiagnosticContext(d: LighthouseDiagnostic): FixSuggestionRequest {
+  const evidence = Array.isArray(d.evidence) ? d.evidence.slice(0, 8) : [];
+  const url = evidence.find((e) => typeof e === 'string' && (e.startsWith('http://') || e.startsWith('https://')));
+  return {
+    source: 'lighthouse',
+    message: d.warning || d.helpText || d.one_line_fix || 'Lighthouse diagnostic',
+    url: typeof url === 'string' ? url : undefined,
+    context: {
+      kind: 'diagnostic',
+      severity: d.severity,
+      one_line_fix: d.one_line_fix,
+      detailed_fix: d.detailed_fix,
+      estimated_impact: d.estimated_impact,
+      lighthouse_audit_id: d.lighthouse_audit_id || d.id,
+      evidence,
+      primary_impact: d.primary_impact,
+    },
+  };
+}
+
+export function buildLighthouseAuditContext(audit: LighthouseAuditRef): FixSuggestionRequest {
+  const items = audit?.details?.items;
+  return {
+    source: 'lighthouse',
+    message: audit.title || audit.id || 'Lighthouse audit',
+    context: {
+      kind: 'audit',
+      id: audit.id,
+      description: audit.description,
+      helpText: audit.helpText,
+      displayValue: audit.displayValue,
+      itemCount: Array.isArray(items) ? items.length : 0,
+    },
+  };
+}
+
+export function buildSecurityFindingContext(f: SecurityFinding): FixSuggestionRequest {
+  return {
+    source: 'security',
+    message: f.message || f.finding_type || 'Security finding',
+    url: f.url,
+    context: {
+      finding_type: f.finding_type,
+      severity: f.severity,
+      recommendation: f.recommendation,
+    },
+  };
+}
+
+export function buildSecurityHeaderContext(label: string, url?: string): FixSuggestionRequest {
+  return {
+    source: 'security',
+    message: `Missing security header: ${label}`,
+    url,
+    context: { kind: 'missing_header', header: label },
+  };
+}
+
+export function buildBrowserErrorContext(row: FlatBrowserErrorRow): FixSuggestionRequest {
+  return {
+    source: 'browser',
+    message: row.message || 'Browser error',
+    url: row.url,
+    context: {
+      kind: row.type,
+      source_url: row.source_url,
+      line: row.line,
+      stack: row.stack ? row.stack.slice(0, 500) : undefined,
+    },
+  };
+}
+
+export function buildBrowserErrorSummaryContext(message: string, sampleUrls: string[], count: number): FixSuggestionRequest {
+  return {
+    source: 'browser',
+    message,
+    url: sampleUrls[0],
+    context: {
+      kind: 'recurring_console',
+      count,
+      sample_urls: sampleUrls.slice(0, 5),
+    },
+  };
+}
+
+export function buildInspectorIssueContext(issue: InspectorIssueRow, pageUrl?: string): FixSuggestionRequest {
+  return {
+    source: 'issue',
+    message: issue.message || 'Page issue',
+    url: pageUrl,
+    context: {
+      severity: issue.severity,
+      type: issue.type,
+      category: issue.category,
+      recommendation: issue.recommendation,
+      detail: issue.detail,
+    },
+  };
+}
+
+export function buildOnPageWarningContext(
+  warning: { message?: string; severity?: string; detail?: string; id?: string },
+  pageUrl: string,
+): FixSuggestionRequest {
+  return {
+    source: 'browser',
+    message: warning.message || 'On-page warning',
+    url: pageUrl,
+    context: {
+      kind: 'on_page_warning',
+      severity: warning.severity,
+      detail: warning.detail,
+      id: warning.id,
+    },
+  };
+}
+
+export function buildCompareIssueContext(row: IssueDeltaRow): FixSuggestionRequest {
+  return {
+    source: 'seo_content',
+    message: row.message || 'Issue change',
+    url: row.url,
+    context: {
+      kind: 'compare_issue',
+      change: row.kind,
+      priority: row.priority,
+      category: row.category,
+    },
+  };
+}
+
+export function buildCompareSecurityContext(row: {
+  message: string;
+  url: string;
+  findingType: string;
+  severity: string;
+  kind: string;
+}): FixSuggestionRequest {
+  return {
+    source: 'security',
+    message: row.message || row.findingType,
+    url: row.url,
+    context: {
+      kind: 'compare_security',
+      change: row.kind,
+      finding_type: row.findingType,
+      severity: row.severity,
+    },
+  };
+}
+
+export function buildMisalignmentContext(item: QueryPageMisalignmentItem): FixSuggestionRequest {
+  return {
+    source: 'seo_content',
+    message: `Query/page misalignment: "${item.keyword}"`,
+    url: item.current_url,
+    context: {
+      kind: 'misalignment',
+      keyword: item.keyword,
+      current_url: item.current_url,
+      suggested_url: item.suggested_url,
+      impressions: item.impressions,
+      position: item.position,
+    },
+  };
+}
+
+export function buildCannibalisationContext(item: CannibalisationItem): FixSuggestionRequest {
+  return {
+    source: 'seo_content',
+    message: `Keyword cannibalisation: "${item.query}"`,
+    context: {
+      kind: 'cannibalisation',
+      query: item.query,
+      pages: (item.pages || []).slice(0, 8).map((p) => ({
+        url: p.url,
+        position: p.position,
+        clicks: p.clicks,
+      })),
+    },
+  };
+}
+
+export function buildKeywordGapContext(row: CompetitorKeywordGapRow): FixSuggestionRequest {
+  return {
+    source: 'seo_content',
+    message: row.keyword || 'Competitor keyword gap',
+    url: row.url ?? undefined,
+    context: {
+      kind: 'keyword_gap',
+      keyword: row.keyword,
+      competitor: row.competitor,
+      volume: row.volume,
+      position: row.position,
+      source: row.source,
+    },
+  };
+}
+
+export function buildRichResultsContext(row: RichResultsValidationRow): FixSuggestionRequest {
+  return {
+    source: 'seo_content',
+    message: row.message || `Rich results ${row.status || 'issue'}`,
+    url: row.url,
+    context: {
+      kind: 'rich_results',
+      status: row.status,
+      provenance: row.provenance,
+    },
+  };
+}
+
+export function buildDuplicateClusterContext(cluster: ContentDuplicateCluster): FixSuggestionRequest {
+  const members = cluster.member_urls || [];
+  return {
+    source: 'seo_content',
+    message: `Duplicate content cluster (${cluster.member_count ?? members.length} URLs)`,
+    url: cluster.representative_url,
+    context: {
+      kind: 'duplicate_cluster',
+      cluster_id: cluster.id,
+      representative_url: cluster.representative_url,
+      member_urls: members.slice(0, 10),
+      member_count: cluster.member_count ?? members.length,
+    },
+  };
+}
+
+export function buildOverviewRecommendationContext(text: string): FixSuggestionRequest {
+  return {
+    source: 'seo_content',
+    message: text,
+    context: { kind: 'overview_recommendation' },
+  };
+}
+
+export function buildRedirectContext(r: ReportRedirect): FixSuggestionRequest {
+  const fromUrl = r.url || r.from;
+  const toUrl = r.final_url || r.to;
+  return {
+    source: 'technical',
+    message: `Redirect ${r.status ?? ''} from ${fromUrl || 'URL'}`,
+    url: fromUrl,
+    context: {
+      kind: 'redirect',
+      status: r.status,
+      final_url: toUrl,
+    },
+  };
+}
+
+export function buildTechnicalLinkIssueContext(message: string, url: string, kind: string): FixSuggestionRequest {
+  return {
+    source: 'technical',
+    message,
+    url,
+    context: { kind },
+  };
+}
+
+export function buildRecommendationBulletContext(text: string, pageUrl?: string): FixSuggestionRequest {
+  return {
+    source: 'issue',
+    message: text,
+    url: pageUrl,
+    context: { kind: 'inspector_recommendation' },
+  };
+}
+
+export function buildContentSignalContext(
+  signal: 'title' | 'meta' | 'h1',
+  label: string,
+  detail: string,
+  statusLabel: string,
+  pageUrl: string,
+): FixSuggestionRequest {
+  return {
+    source: 'seo_content',
+    message: `${label}: ${statusLabel}`,
+    url: pageUrl,
+    context: {
+      kind: 'content_signal',
+      signal,
+      detail,
+      status: statusLabel,
+    },
+  };
+}
+
+export function buildLighthouseFailureContext(helpText: string, id?: string, pageUrl?: string): FixSuggestionRequest {
+  return {
+    source: 'lighthouse',
+    message: helpText || id || 'Lighthouse failure',
+    url: pageUrl,
+    context: { kind: 'top_failure', id: id || helpText },
+  };
+}
diff --git a/web/src/lib/formatPipelineLog.test.ts b/web/src/lib/formatPipelineLog.test.ts
new file mode 100644
index 00000000..b0b5ee61
--- /dev/null
+++ b/web/src/lib/formatPipelineLog.test.ts
@@ -0,0 +1,78 @@
+import { describe, expect, it } from 'vitest';
+import {
+  computeEta,
+  extractLatestProgress,
+  formatDurationMs,
+  parsePipelineLog,
+  parsePipelineProgressEvents,
+  resolveActiveProgress,
+  stepLabel,
+} from '@/lib/formatPipelineLog';
+
+describe('parsePipelineProgressEvents', () => {
+  it('parses @progress JSON lines', () => {
+    const raw = [
+      '[Crawl] Starting...',
+      '@progress {"phase":"crawl","step":"fetch","current":3,"total":10,"ts":1000,"elapsed_ms":3000,"avg_ms":1000}',
+      '@progress {"phase":"crawl","step":"fetch","current":6,"total":10,"ts":2000,"elapsed_ms":6000,"avg_ms":1000}',
+    ].join('\n');
+    const events = parsePipelineProgressEvents(raw);
+    expect(events).toHaveLength(2);
+    expect(events[1]?.current).toBe(6);
+    expect(extractLatestProgress(events)?.total).toBe(10);
+  });
+
+  it('renders @progress lines as rich log entries', () => {
+    const raw = [
+      '@progress {"phase":"crawl","step":"start","message":"Crawling pages","ts":1}',
+      '@progress {"phase":"crawl","step":"fetch","current":5,"total":30,"url":"https://example.com/about","ts":2}',
+      '@progress {"phase":"report","step":"categories","message":"Building categories","ts":3}',
+      '[Report] Done.',
+    ].join('\n');
+    const lines = parsePipelineLog(raw);
+    expect(lines.some((l) => l.text.startsWith('@progress'))).toBe(false);
+    expect(lines.some((l) => l.kind === 'activity' && l.text.includes('example.com/about'))).toBe(true);
+    expect(lines.some((l) => l.text.includes('[Crawl]'))).toBe(true);
+    expect(lines.some((l) => l.text.includes('[Report]'))).toBe(true);
+  });
+});
+
+describe('resolveActiveProgress', () => {
+  it('marks plot complete when job succeeded but last event was plot start', () => {
+    const events = [
+      { phase: 'crawl' as const, step: 'done', ts: 1, message: 'crawl complete' },
+      { phase: 'plot' as const, step: 'start', ts: 2, message: 'plot starting' },
+    ];
+    const resolved = resolveActiveProgress(events, 'success');
+    expect(resolved?.step).toBe('done');
+    expect(resolved?.phase).toBe('plot');
+  });
+});
+
+describe('computeEta', () => {
+  it('estimates remaining time from avg_ms', () => {
+    const history = [
+      { phase: 'crawl' as const, step: 'fetch', ts: 1, current: 2, total: 10, elapsed_ms: 2000, avg_ms: 1000 },
+      { phase: 'crawl' as const, step: 'fetch', ts: 2, current: 5, total: 10, elapsed_ms: 5000, avg_ms: 1000 },
+    ];
+    const latest = history[1]!;
+    const eta = computeEta(latest, history);
+    expect(eta.percent).toBe(50);
+    expect(eta.ratePerSec).toBeCloseTo(1, 1);
+    expect(eta.remainingMs).toBe(5000);
+  });
+});
+
+describe('stepLabel', () => {
+  it('prefers message over step map', () => {
+    expect(stepLabel('fetch', 'Custom step')).toBe('Custom step');
+    expect(stepLabel('categories')).toBe('Building categories');
+  });
+});
+
+describe('formatDurationMs', () => {
+  it('formats seconds and minutes', () => {
+    expect(formatDurationMs(45000)).toBe('45s');
+    expect(formatDurationMs(125000)).toBe('2m 5s');
+  });
+});
diff --git a/web/src/lib/formatPipelineLog.ts b/web/src/lib/formatPipelineLog.ts
index 4cf5ea1c..5f004725 100644
--- a/web/src/lib/formatPipelineLog.ts
+++ b/web/src/lib/formatPipelineLog.ts
@@ -1,6 +1,7 @@
 export type PipelineLogLineKind =
   | 'section'
   | 'progress'
+  | 'activity'
   | 'info'
   | 'success'
   | 'warning'
@@ -9,7 +10,38 @@ export type PipelineLogLineKind =
   | 'noise'
   | 'plain';
 
-export type PipelinePhase = 'config' | 'crawl' | 'lighthouse' | 'report' | 'plot' | 'other';
+export type PipelinePhase =
+  | 'config'
+  | 'crawl'
+  | 'lighthouse'
+  | 'report'
+  | 'keywords'
+  | 'plot'
+  | 'optional'
+  | 'other';
+
+export type ProgressPhase = PipelinePhase;
+
+export interface PipelineProgressEvent {
+  phase: ProgressPhase;
+  step: string;
+  ts: number;
+  current?: number;
+  total?: number;
+  url?: string;
+  message?: string;
+  elapsed_ms?: number;
+  avg_ms?: number;
+}
+
+export interface PipelineEtaResult {
+  remainingMs: number | null;
+  ratePerSec: number | null;
+  elapsedMs: number | null;
+  percent: number | null;
+}
+
+const PROGRESS_PREFIX = '@progress ';
 
 export interface PipelineLogLine {
   id: number;
@@ -17,6 +49,8 @@ export interface PipelineLogLine {
   kind: PipelineLogLineKind;
   phase: PipelinePhase;
   progress?: { percent: number; current: number; total: number };
+  /** Set when line was derived from a structured @progress event. */
+  progressEvent?: PipelineProgressEvent;
 }
 
 export interface PipelineLogGroup {
@@ -28,15 +62,47 @@ export interface PipelineLogGroup {
 
 const ANSI_RE = /\x1b\[[0-9;]*m/g;
 
-const PHASE_LABELS: Record<PipelinePhase, string> = {
+export const PHASE_LABELS: Record<PipelinePhase, string> = {
   config: 'Settings',
   crawl: 'Crawl',
   lighthouse: 'Lighthouse',
   report: 'Site audit',
+  keywords: 'Keywords',
   plot: 'Charts',
+  optional: 'Optional audits',
   other: 'Output',
 };
 
+const STEP_LABELS: Record<string, string> = {
+  start: 'Starting',
+  done: 'Complete',
+  fetch: 'Fetching pages',
+  audit: 'Running audits',
+  load_crawl: 'Loading crawl data',
+  load_data: 'Loading crawl data',
+  build_edges: 'Building link graph',
+  write_db: 'Saving link graph',
+  link_edges: 'Loading link edges',
+  seo_summary: 'SEO summary',
+  site_level: 'Site-level checks',
+  contact_intelligence: 'Contact intelligence',
+  subdomains: 'Subdomain inventory',
+  security_scan: 'Security scan',
+  content_analysis: 'Content analysis',
+  categories: 'Building categories',
+  content_analytics: 'Content analytics',
+  write_payload: 'Saving report',
+};
+
+export const PIPELINE_STEPPER_PHASES: ProgressPhase[] = [
+  'config',
+  'crawl',
+  'lighthouse',
+  'report',
+  'keywords',
+  'plot',
+];
+
 /** Strip ANSI escapes and tqdm control chars. */
 export function stripAnsi(raw: string): string {
   return raw.replace(ANSI_RE, '').replace(/\x1b\].*?\x07/g, '');
@@ -55,6 +121,7 @@ export function phaseFromSectionTag(tag: string): PipelinePhase {
 function classifyLine(line: string): PipelineLogLineKind {
   const t = line.trim();
   if (!t) return 'plain';
+  if (t.startsWith(PROGRESS_PREFIX)) return 'noise';
   if (/^Exception ignored while calling deallocator/i.test(t)) return 'noise';
   if (/PythonFinalizationError|cannot join thread at interpreter shutdown/i.test(t)) {
     return 'noise';
@@ -136,6 +203,211 @@ function collapseConsecutiveProgress(lines: PipelineLogLine[]): PipelineLogLine[
   return out;
 }
 
+function parseProgressPayload(line: string): PipelineProgressEvent | null {
+  const t = line.trim();
+  if (!t.startsWith(PROGRESS_PREFIX)) return null;
+  try {
+    const parsed = JSON.parse(t.slice(PROGRESS_PREFIX.length)) as PipelineProgressEvent;
+    if (!parsed || typeof parsed.phase !== 'string' || typeof parsed.step !== 'string') {
+      return null;
+    }
+    return parsed;
+  } catch {
+    return null;
+  }
+}
+
+/** Extract structured @progress events from raw log text. */
+export function parsePipelineProgressEvents(raw: string): PipelineProgressEvent[] {
+  const cleaned = stripAnsi(raw);
+  const events: PipelineProgressEvent[] = [];
+  for (const physical of cleaned.split('\n')) {
+    const segments = physical.split('\r');
+    const line = (segments[segments.length - 1] ?? '').trim();
+    const evt = parseProgressPayload(line);
+    if (evt) events.push(evt);
+  }
+  return events;
+}
+
+export function extractLatestProgress(events: PipelineProgressEvent[]): PipelineProgressEvent | null {
+  if (!events.length) return null;
+  return events[events.length - 1] ?? null;
+}
+
+export type PipelineJobStatus = 'starting' | 'running' | 'success' | 'error' | '';
+
+/** Pick the progress snapshot to show in the header (respects finished jobs). */
+export function resolveActiveProgress(
+  events: PipelineProgressEvent[],
+  jobStatus: PipelineJobStatus = '',
+): PipelineProgressEvent | null {
+  if (!events.length) return null;
+  const latest = events[events.length - 1]!;
+  if (jobStatus === 'success') {
+    if (latest.step === 'start') {
+      return {
+        ...latest,
+        step: 'done',
+        message: latest.message?.replace(/\bstarting\b/i, 'complete') ?? `${latest.phase} complete`,
+      };
+    }
+    const lastDone = [...events].reverse().find((e) => e.step === 'done');
+    if (lastDone) return lastDone;
+  }
+  if (jobStatus === 'error' && latest.step === 'start') {
+    const lastDone = [...events].reverse().find((e) => e.step === 'done' && e.phase === latest.phase);
+    if (lastDone) return lastDone;
+  }
+  return latest;
+}
+
+function phaseFromProgressEvent(phase: string): PipelinePhase {
+  const p = phase.toLowerCase();
+  if (p === 'config' || p === 'crawl' || p === 'lighthouse' || p === 'report' || p === 'keywords' || p === 'plot') {
+    return p;
+  }
+  if (p === 'optional') return 'optional';
+  return 'other';
+}
+
+function progressEventToLogLine(evt: PipelineProgressEvent, id: number): PipelineLogLine {
+  const phase = phaseFromProgressEvent(evt.phase);
+  const label = PHASE_LABELS[phase] ?? evt.phase;
+  const stepText = stepLabel(evt.step, evt.message);
+
+  if (evt.step === 'start') {
+    return {
+      id,
+      text: `[${label}] ${stepText}`,
+      kind: 'section',
+      phase,
+      progressEvent: evt,
+    };
+  }
+
+  if (evt.step === 'done') {
+    return {
+      id,
+      text: `[${label}] ${stepText}`,
+      kind: 'success',
+      phase,
+      progressEvent: evt,
+    };
+  }
+
+  const hasCounts =
+    evt.current != null && evt.total != null && evt.total > 0;
+  const progress = hasCounts
+    ? {
+        percent: Math.min(100, Math.round(((evt.current ?? 0) / (evt.total ?? 1)) * 100)),
+        current: evt.current ?? 0,
+        total: evt.total ?? 0,
+      }
+    : undefined;
+
+  if (evt.step === 'fetch' && evt.url) {
+    const countSuffix = hasCounts ? ` (${evt.current}/${evt.total})` : '';
+    return {
+      id,
+      text: `→ ${evt.url}${countSuffix}`,
+      kind: 'activity',
+      phase,
+      progress,
+      progressEvent: evt,
+    };
+  }
+
+  if (hasCounts) {
+    return {
+      id,
+      text: `${stepText} ${evt.current}/${evt.total}`,
+      kind: 'progress',
+      phase,
+      progress,
+      progressEvent: evt,
+    };
+  }
+
+  return {
+    id,
+    text: `[${label}] ${stepText}`,
+    kind: 'info',
+    phase,
+    progressEvent: evt,
+  };
+}
+
+
+export function stepLabel(step: string, message?: string): string {
+  if (message?.trim()) return message.trim();
+  return STEP_LABELS[step] ?? step.replace(/_/g, ' ');
+}
+
+/** Rolling ETA from progress event history (last 10 samples with current+total). */
+export function computeEta(
+  latest: PipelineProgressEvent | null,
+  history: PipelineProgressEvent[],
+): PipelineEtaResult {
+  if (!latest) {
+    return { remainingMs: null, ratePerSec: null, elapsedMs: null, percent: null };
+  }
+  const elapsedMs = latest.elapsed_ms ?? null;
+  let percent: number | null = null;
+  if (
+    latest.current != null &&
+    latest.total != null &&
+    latest.total > 0
+  ) {
+    percent = Math.min(100, Math.round((latest.current / latest.total) * 100));
+  }
+
+  const samples = history.filter(
+    (e) =>
+      e.phase === latest.phase &&
+      e.step === latest.step &&
+      e.current != null &&
+      e.total != null &&
+      e.total > 0 &&
+      e.elapsed_ms != null,
+  );
+  const tail = samples.slice(-10);
+  let ratePerSec: number | null = null;
+  if (latest.avg_ms != null && latest.avg_ms > 0) {
+    ratePerSec = 1000 / latest.avg_ms;
+  } else if (tail.length >= 2) {
+    const first = tail[0];
+    const last = tail[tail.length - 1];
+    const deltaCurrent = (last.current ?? 0) - (first.current ?? 0);
+    const deltaMs = (last.elapsed_ms ?? 0) - (first.elapsed_ms ?? 0);
+    if (deltaCurrent > 0 && deltaMs > 0) {
+      ratePerSec = (deltaCurrent / deltaMs) * 1000;
+    }
+  }
+
+  let remainingMs: number | null = null;
+  if (
+    latest.current != null &&
+    latest.total != null &&
+    latest.total > latest.current &&
+    ratePerSec != null &&
+    ratePerSec > 0
+  ) {
+    remainingMs = Math.round(((latest.total - latest.current) / ratePerSec) * 1000);
+  }
+
+  return { remainingMs, ratePerSec, elapsedMs, percent };
+}
+
+export function formatDurationMs(ms: number | null): string {
+  if (ms == null || !Number.isFinite(ms) || ms < 0) return '—';
+  const sec = Math.round(ms / 1000);
+  if (sec < 60) return `${sec}s`;
+  const min = Math.floor(sec / 60);
+  const rem = sec % 60;
+  return rem ? `${min}m ${rem}s` : `${min}m`;
+}
+
 /** Split raw job log into display lines (collapse tqdm spam, tag phases, fold shutdown noise). */
 export function parsePipelineLog(raw: string): PipelineLogLine[] {
   const cleaned = stripAnsi(raw);
@@ -148,6 +420,13 @@ export function parsePipelineLog(raw: string): PipelineLogLine[] {
     const line = (segments[segments.length - 1] ?? '').replace(/\s+$/, '');
     if (!line.trim()) continue;
 
+    const trimmed = line.trim();
+    const progressEvt = parseProgressPayload(trimmed);
+    if (progressEvt) {
+      parsed.push(progressEventToLogLine(progressEvt, id++));
+      continue;
+    }
+
     const kind = classifyLine(line);
     const entry: PipelineLogLine = { id: id++, text: line, kind, phase: 'other' };
     if (kind === 'progress') {
@@ -212,6 +491,7 @@ export function getPipelineLogStats(lines: PipelineLogLine[]) {
 export const PIPELINE_LOG_LINE_CLASS: Record<PipelineLogLineKind, string> = {
   section: 'text-sky-300',
   progress: 'text-muted-foreground',
+  activity: 'text-[#e6edf3]/90',
   info: 'text-violet-300',
   success: 'text-green-400',
   warning: 'text-amber-300',
@@ -226,6 +506,8 @@ export const PHASE_CHIP_CLASS: Record<PipelinePhase, string> = {
   crawl: 'border-blue-500/40 bg-blue-500/10 text-blue-200',
   lighthouse: 'border-amber-500/40 bg-amber-500/10 text-amber-200',
   report: 'border-green-500/40 bg-green-500/10 text-green-200',
+  keywords: 'border-pink-500/40 bg-pink-500/10 text-pink-200',
   plot: 'border-cyan-500/40 bg-cyan-500/10 text-cyan-200',
+  optional: 'border-orange-500/40 bg-orange-500/10 text-orange-200',
   other: 'border-default bg-brand-800/60 text-muted-foreground',
 };
diff --git a/web/src/lib/llmConfigSchema.ts b/web/src/lib/llmConfigSchema.ts
index 104b2611..b63f59ce 100644
--- a/web/src/lib/llmConfigSchema.ts
+++ b/web/src/lib/llmConfigSchema.ts
@@ -63,10 +63,10 @@ export const LLM_CONFIG_SECTIONS = [
       { key: 'llm_enable_keyword_clusters', label: 'Topic clusters (AI)', type: 'bool', defaultValue: true },
       {
         key: 'llm_enable_issue_fixes',
-        label: 'Issue fix suggestions',
+        label: 'AI fix suggestions',
         type: 'bool',
         defaultValue: true,
-        help: 'Generate AI fix text for top traffic-weighted audit issues at report build time.',
+        help: 'On-demand AI fix buttons across Issues, Lighthouse, Security, and other audit surfaces; also enriches top issues at report build.',
       },
       {
         key: 'llm_enable_audit_summary',
diff --git a/web/src/lib/pipelineConfigSchema.ts b/web/src/lib/pipelineConfigSchema.ts
index 2d348a50..0c46658c 100644
--- a/web/src/lib/pipelineConfigSchema.ts
+++ b/web/src/lib/pipelineConfigSchema.ts
@@ -56,6 +56,87 @@ export const PIPELINE_CONFIG_SECTIONS: PipelineConfigSection[] = [
         placeholder: 'https://example.com',
         help: 'Required before running a crawl or audit. Enter the property site to analyze.',
       },
+      {
+        key: 'crawl_discovery_mode',
+        label: 'URL discovery mode',
+        type: 'select',
+        defaultValue: 'spider',
+        options: [
+          { value: 'spider', label: 'Spider — follow internal links' },
+          { value: 'list', label: 'List only — audit pasted/uploaded URLs' },
+          { value: 'sitemap', label: 'Sitemap only — no link following' },
+          { value: 'hybrid', label: 'Hybrid — spider + sitemap + list' },
+        ],
+        help: 'Controls how URLs enter the crawl queue. List mode does not follow links.',
+      },
+      {
+        key: 'crawl_url_list',
+        label: 'URL list',
+        type: 'textarea',
+        defaultValue: '',
+        span: 2,
+        placeholder: 'https://example.com/page-a\nhttps://example.com/page-b',
+        help: 'One URL per line (or comma-separated). Required for List mode; optional for Hybrid.',
+        visibleWhen: { key: 'crawl_discovery_mode', not: ['spider', 'sitemap'] },
+      },
+      {
+        key: 'crawl_user_agent_preset',
+        label: 'Crawl user-agent preset',
+        type: 'select',
+        defaultValue: 'default',
+        options: [
+          { value: 'default', label: 'Default crawler UA' },
+          { value: 'mobile', label: 'Mobile Safari' },
+          { value: 'custom', label: 'Custom UA string' },
+        ],
+      },
+      {
+        key: 'crawl_user_agent_custom',
+        label: 'Custom user-agent',
+        type: 'text',
+        defaultValue: '',
+        visibleWhen: { key: 'crawl_user_agent_preset', not: ['default', 'mobile'] },
+      },
+      {
+        key: 'crawl_auth_username',
+        label: 'HTTP Basic auth username',
+        type: 'text',
+        defaultValue: '',
+      },
+      {
+        key: 'crawl_auth_password',
+        label: 'HTTP Basic auth password',
+        type: 'text',
+        defaultValue: '',
+      },
+      {
+        key: 'crawl_extra_headers',
+        label: 'Extra request headers',
+        type: 'textarea',
+        defaultValue: '',
+        help: 'One header per line: Name: value',
+      },
+      {
+        key: 'crawl_cookies',
+        label: 'Cookie header value',
+        type: 'text',
+        defaultValue: '',
+      },
+      {
+        key: 'crawl_robots_txt_override',
+        label: 'Robots.txt override',
+        type: 'textarea',
+        defaultValue: '',
+        help: 'Paste robots.txt rules for staging crawls instead of fetching live robots.txt.',
+      },
+      {
+        key: 'custom_extractors',
+        label: 'Custom extractors (JSON)',
+        type: 'textarea',
+        defaultValue: '',
+        span: 2,
+        help: 'JSON array: [{\"name\":\"sku\",\"type\":\"css\",\"selector\":\"[data-sku]\",\"attr\":\"data-sku\"}]',
+      },
       { key: 'max_pages', label: 'Crawl limit (URLs)', type: 'number', defaultValue: '500' },
       { key: 'concurrency', label: 'Concurrent requests', type: 'number', defaultValue: '8' },
       { key: 'timeout', label: 'Timeout (s)', type: 'number', defaultValue: '12' },
@@ -255,6 +336,27 @@ export const PIPELINE_CONFIG_SECTIONS: PipelineConfigSection[] = [
       { key: 'image_probe_concurrency', label: 'Image probe concurrency', type: 'number', defaultValue: '6' },
       { key: 'image_probe_timeout', label: 'Image probe timeout (seconds)', type: 'number', defaultValue: '8' },
       { key: 'image_unoptimized_min_kb', label: 'Unoptimized image min size (KB)', type: 'number', defaultValue: '200' },
+      {
+        key: 'enable_subdomain_discovery',
+        label: 'Subdomain discovery',
+        type: 'bool',
+        defaultValue: true,
+        help: 'Build passive subdomain inventory from crawl, GSC, and certificate transparency.',
+      },
+      {
+        key: 'subdomain_ct_lookup',
+        label: 'Certificate transparency lookup',
+        type: 'bool',
+        defaultValue: true,
+        help: 'Query crt.sh when subdomain discovery is enabled.',
+      },
+      {
+        key: 'enable_rdap_org_lookup',
+        label: 'RDAP organization lookup',
+        type: 'bool',
+        defaultValue: true,
+        help: 'Fetch registrant org name for contact intelligence (best-effort).',
+      },
     ],
   },
   {
@@ -309,6 +411,26 @@ export const PIPELINE_CONFIG_SECTIONS: PipelineConfigSection[] = [
       { key: 'run_lighthouse', label: 'Run single-URL Lighthouse', type: 'bool', defaultValue: true },
       { key: 'run_lighthouse_on_pages', label: 'Run Lighthouse on crawled pages', type: 'bool', defaultValue: true },
       { key: 'enable_crux', label: 'Fetch CrUX field Core Web Vitals', type: 'bool', defaultValue: false, help: 'Uses Chrome UX Report API for real-user LCP, INP, CLS at origin level.' },
+      { key: 'enable_rich_results_validation', label: 'Validate structured data (Rich Results)', type: 'bool', defaultValue: false },
+      {
+        key: 'google_rich_results_api_key',
+        label: 'Google Rich Results API key',
+        type: 'text',
+        defaultValue: '',
+        help: 'Optional API key for Rich Results Test API when GSC OAuth is unavailable.',
+        placeholder: 'AIza…',
+      },
+      {
+        key: 'enable_axe',
+        label: 'Run axe accessibility scan (browser crawl)',
+        type: 'bool',
+        defaultValue: false,
+        help: 'Requires crawl_render_mode javascript or auto. No-op on static-only crawls.',
+      },
+      { key: 'enable_spell_check', label: 'Spell-check page excerpts', type: 'bool', defaultValue: false },
+      { key: 'enable_html_validation', label: 'Basic HTML validation warnings', type: 'bool', defaultValue: false },
+      { key: 'enable_amp_audit', label: 'AMP canonical pairing audit', type: 'bool', defaultValue: false },
+      { key: 'enable_wayback_lookup', label: 'Wayback lookup for 404 URLs', type: 'bool', defaultValue: false },
       { key: 'lighthouse_max_pages', label: 'Performance sample size (URLs)', type: 'number', defaultValue: '2' },
       { key: 'lighthouse_concurrency', label: 'Lighthouse parallel URLs', type: 'number', defaultValue: '2' },
     ],
@@ -631,6 +753,11 @@ export function validatePipelineRun({
   ) {
     errors.push(browserStatus.message?.trim() || BROWSER_CRAWL_UNAVAILABLE_MSG);
   }
+  const discovery = String(state?.crawl_discovery_mode ?? 'spider').trim().toLowerCase();
+  const urlList = String(state?.crawl_url_list ?? '').trim();
+  if (runIncludesCrawl(state, command) && discovery === 'list' && !urlList) {
+    errors.push('URL list is required when discovery mode is List. Paste URLs in Audit settings.');
+  }
   return errors;
 }
 
diff --git a/web/src/lib/pipelineLiveEstimate.test.ts b/web/src/lib/pipelineLiveEstimate.test.ts
new file mode 100644
index 00000000..2c6cd411
--- /dev/null
+++ b/web/src/lib/pipelineLiveEstimate.test.ts
@@ -0,0 +1,92 @@
+import { describe, expect, it } from 'vitest';
+import { buildInitialPipelineConfigState } from '@/lib/pipelineConfigSchema';
+import type { PipelineProgressEvent } from '@/lib/formatPipelineLog';
+import { buildPipelineRunPreview } from '@/lib/pipelineRunPreview';
+import {
+  computeLivePipelineEstimate,
+  formatLivePipelineDuration,
+  observedCrawlPages,
+} from '@/lib/pipelineLiveEstimate';
+
+function crawlFetch(current: number, total: number, ts: number, avgMs = 1200): PipelineProgressEvent {
+  return {
+    phase: 'crawl',
+    step: 'fetch',
+    ts,
+    current,
+    total,
+    elapsed_ms: current * avgMs,
+    avg_ms: avgMs,
+    url: `https://example.com/p${current}`,
+  };
+}
+
+describe('computeLivePipelineEstimate', () => {
+  const preview = buildPipelineRunPreview({
+    presetId: 'full-audit',
+    configState: { ...buildInitialPipelineConfigState(), max_pages: '30', lighthouse_max_pages: '2' },
+  });
+
+  it('reads observed crawl pages from progress stream', () => {
+    const events = [
+      { phase: 'crawl' as const, step: 'start', ts: 1000 },
+      crawlFetch(5, 30, 5000),
+      crawlFetch(10, 30, 10000),
+    ];
+    expect(observedCrawlPages(events)).toBe(10);
+  });
+
+  it('shrinks remaining time as crawl progresses', () => {
+    const early = computeLivePipelineEstimate(
+      preview,
+      [
+        { phase: 'crawl', step: 'start', ts: 0 },
+        crawlFetch(3, 30, 4000, 1500),
+      ],
+      crawlFetch(3, 30, 4000, 1500),
+      'running',
+    );
+    const later = computeLivePipelineEstimate(
+      preview,
+      [
+        { phase: 'crawl', step: 'start', ts: 0 },
+        crawlFetch(20, 30, 30000, 1000),
+      ],
+      crawlFetch(20, 30, 30000, 1000),
+      'running',
+    );
+    expect(early?.remainingMs).toBeTruthy();
+    expect(later?.remainingMs).toBeTruthy();
+    expect(later!.remainingMs!).toBeLessThan(early!.remainingMs!);
+  });
+
+  it('returns zero remaining when job succeeded', () => {
+    const events: PipelineProgressEvent[] = [
+      { phase: 'crawl', step: 'start', ts: 0 },
+      crawlFetch(30, 30, 20000),
+      { phase: 'crawl', step: 'done', ts: 25000 },
+      { phase: 'plot', step: 'done', ts: 90000 },
+    ];
+    const estimate = computeLivePipelineEstimate(
+      preview,
+      events,
+      events[events.length - 1]!,
+      'success',
+    );
+    expect(estimate?.remainingMs).toBe(0);
+    expect(estimate?.elapsedMs).toBeGreaterThan(0);
+  });
+
+  it('formats elapsed and remaining for the UI', () => {
+    const text = formatLivePipelineDuration({
+      elapsedMs: 125_000,
+      remainingMs: 95_000,
+      totalMs: 220_000,
+      observedCrawlPages: 12,
+      ratePerSec: 1.2,
+      source: 'live',
+    });
+    expect(text).toContain('elapsed');
+    expect(text).toContain('left');
+  });
+});
diff --git a/web/src/lib/pipelineLiveEstimate.ts b/web/src/lib/pipelineLiveEstimate.ts
new file mode 100644
index 00000000..1a2dc2fe
--- /dev/null
+++ b/web/src/lib/pipelineLiveEstimate.ts
@@ -0,0 +1,181 @@
+import {
+  computeEta,
+  formatDurationMs,
+  type PipelineJobStatus,
+  type PipelineProgressEvent,
+  type ProgressPhase,
+} from '@/lib/formatPipelineLog';
+import type { PipelinePhaseTiming, PipelineRunPreview } from '@/lib/pipelineRunPreview';
+
+export interface LivePipelineEstimate {
+  /** Wall-clock elapsed since first progress event. */
+  elapsedMs: number;
+  /** Projected time left for this run (ms). */
+  remainingMs: number | null;
+  /** elapsed + remaining when known. */
+  totalMs: number | null;
+  /** Pages crawled so far from live @progress samples. */
+  observedCrawlPages: number | null;
+  /** Rolling throughput for the active counted step (pages/sec). */
+  ratePerSec: number | null;
+  source: 'live' | 'static';
+}
+
+const PIPELINE_PHASE_ORDER: ProgressPhase[] = [
+  'config',
+  'crawl',
+  'lighthouse',
+  'report',
+  'keywords',
+  'plot',
+];
+
+export function observedCrawlPages(events: PipelineProgressEvent[]): number | null {
+  const fetchEvents = events.filter(
+    (e) => e.phase === 'crawl' && e.step === 'fetch' && e.current != null && e.current > 0,
+  );
+  if (!fetchEvents.length) return null;
+  return fetchEvents[fetchEvents.length - 1]!.current ?? null;
+}
+
+function phaseIsDone(phase: ProgressPhase, events: PipelineProgressEvent[]): boolean {
+  return events.some((e) => e.phase === phase && e.step === 'done');
+}
+
+function scalePendingPhases(
+  typicalSeconds: number,
+  preview: PipelineRunPreview,
+  observedPages: number | null,
+): number {
+  if (observedPages == null || preview.maxCrawlPages == null || preview.typicalCrawlPages == null) {
+    return typicalSeconds;
+  }
+  const baseline = Math.max(1, preview.typicalCrawlPages);
+  const ratio = observedPages / baseline;
+  const clamped = Math.max(0.35, Math.min(1.25, ratio));
+  return typicalSeconds * clamped;
+}
+
+function pendingPhaseSeconds(
+  timing: PipelinePhaseTiming,
+  preview: PipelineRunPreview,
+  observedPages: number | null,
+): number {
+  if (timing.phase === 'report' && observedPages != null) {
+    const base = 25 + observedPages * 0.012;
+    return Math.max(15, base);
+  }
+  if (timing.phase === 'lighthouse' && observedPages != null && preview.lighthousePages != null) {
+    const sample = Math.min(preview.lighthousePages, observedPages);
+    if (sample <= 0) return timing.typicalSeconds;
+    const workers = 2;
+    return 12 + (sample / workers) * 38 * 0.92;
+  }
+  return scalePendingPhases(timing.typicalSeconds, preview, observedPages);
+}
+
+/**
+ * Rolling pipeline ETA using live @progress averages for the active step
+ * and scaled static estimates for phases still ahead.
+ */
+export function computeLivePipelineEstimate(
+  preview: PipelineRunPreview,
+  events: PipelineProgressEvent[],
+  latest: PipelineProgressEvent | null,
+  jobStatus: PipelineJobStatus = '',
+): LivePipelineEstimate | null {
+  if (!events.length) return null;
+
+  const observed = observedCrawlPages(events);
+  const pipelineStart = events[0]?.ts;
+  const elapsedMs =
+    pipelineStart != null && latest?.ts != null
+      ? Math.max(0, latest.ts - pipelineStart)
+      : Math.max(0, latest?.elapsed_ms ?? 0);
+
+  if (jobStatus === 'success' || jobStatus === 'error') {
+    const totalMs = pipelineStart != null && events.length > 1
+      ? Math.max(elapsedMs, (events[events.length - 1]!.ts ?? 0) - pipelineStart)
+      : elapsedMs;
+    return {
+      elapsedMs: totalMs,
+      remainingMs: 0,
+      totalMs,
+      observedCrawlPages: observed,
+      ratePerSec: null,
+      source: 'live',
+    };
+  }
+
+  if (!latest) {
+    return {
+      elapsedMs,
+      remainingMs: preview.timeMaxSeconds * 1000,
+      totalMs: elapsedMs + preview.timeMaxSeconds * 1000,
+      observedCrawlPages: observed,
+      ratePerSec: null,
+      source: 'static',
+    };
+  }
+
+  const phaseEta = computeEta(latest, events);
+  const activePhase = latest.phase as ProgressPhase;
+  const activeIdx = PIPELINE_PHASE_ORDER.indexOf(activePhase);
+
+  let remainingMs = 0;
+
+  for (const timing of preview.phaseTimings) {
+    const phase = timing.phase as ProgressPhase;
+    const phaseIdx = PIPELINE_PHASE_ORDER.indexOf(phase);
+    if (phaseIdx < 0) continue;
+    if (phaseIsDone(phase, events)) continue;
+    if (activeIdx >= 0 && phaseIdx < activeIdx) continue;
+
+    const isActive = phase === activePhase && latest.step !== 'done';
+
+    if (isActive) {
+      if (phaseEta.remainingMs != null && phaseEta.remainingMs > 0) {
+        remainingMs += phaseEta.remainingMs;
+      } else if (timing.typicalSeconds > 0) {
+        const pct = phaseEta.percent ?? 0;
+        const fallback = timing.typicalSeconds * 1000;
+        remainingMs += fallback * Math.max(0.08, (100 - pct) / 100);
+      }
+      continue;
+    }
+
+    if (activeIdx >= 0 && phaseIdx > activeIdx) {
+      remainingMs += pendingPhaseSeconds(timing, preview, observed) * 1000;
+    }
+  }
+
+  if (remainingMs <= 0 && latest.step === 'start') {
+    const timing = preview.phaseTimings.find((t) => t.phase === activePhase);
+    if (timing) remainingMs = timing.typicalSeconds * 1000;
+  }
+
+  remainingMs = Math.round(Math.max(0, remainingMs));
+
+  return {
+    elapsedMs,
+    remainingMs,
+    totalMs: elapsedMs + remainingMs,
+    observedCrawlPages: observed,
+    ratePerSec: phaseEta.ratePerSec,
+    source: 'live',
+  };
+}
+
+export function formatLivePipelineDuration(estimate: LivePipelineEstimate): string {
+  const elapsed = formatDurationMs(estimate.elapsedMs);
+  if (estimate.remainingMs == null || estimate.remainingMs <= 0) {
+    return elapsed;
+  }
+  const remaining = formatDurationMs(estimate.remainingMs);
+  const total =
+    estimate.totalMs != null ? formatDurationMs(estimate.totalMs) : null;
+  if (total) {
+    return `${elapsed} elapsed · ~${remaining} left (~${total} total)`;
+  }
+  return `${elapsed} elapsed · ~${remaining} left`;
+}
diff --git a/web/src/lib/pipelineRunPreview.test.ts b/web/src/lib/pipelineRunPreview.test.ts
index 1158b57c..630a85a3 100644
--- a/web/src/lib/pipelineRunPreview.test.ts
+++ b/web/src/lib/pipelineRunPreview.test.ts
@@ -1,6 +1,7 @@
 import { describe, expect, it } from 'vitest';
 import { buildInitialPipelineConfigState } from '@/lib/pipelineConfigSchema';
-import { buildPipelineRunPreview } from '@/lib/pipelineRunPreview';
+import { buildPipelineRunPreview, formatPipelineRunDuration } from '@/lib/pipelineRunPreview';
+import { applyCrawlPreset } from '@/lib/crawlPresets';
 
 describe('buildPipelineRunPreview', () => {
   it('includes crawl and report for full audit', () => {
@@ -23,15 +24,42 @@ describe('buildPipelineRunPreview', () => {
     expect(preview.phases.some((p) => p.id === 'report')).toBe(false);
   });
 
-  it('reflects crawl preset max pages', () => {
+  it('applies crawl preset patch when computing preview', () => {
     const preview = buildPipelineRunPreview({
       presetId: 'full-audit',
-      configState: { ...buildInitialPipelineConfigState(), max_pages: '2000' },
+      configState: buildInitialPipelineConfigState(),
       crawlPresetId: 'spa',
     });
-    expect(preview.maxCrawlPages).toBe(2000);
+    const patched = applyCrawlPreset('spa', buildInitialPipelineConfigState());
+    expect(preview.maxCrawlPages).toBe(Number(patched.max_pages));
     expect(preview.configRows.some((r) => r.label === 'Crawl preset' && r.value === 'SPA / JavaScript')).toBe(
       true,
     );
   });
+
+  it('does not assume every page in the crawl limit will be fetched', () => {
+    const performance = applyCrawlPreset('performance', buildInitialPipelineConfigState());
+    const preview = buildPipelineRunPreview({
+      presetId: 'full-audit',
+      configState: performance,
+      crawlPresetId: 'performance',
+    });
+    expect(formatPipelineRunDuration(preview.timeMinSeconds, preview.timeMaxSeconds)).not.toBe(
+      '12 min – 21 min',
+    );
+    expect(preview.timeMinSeconds).toBeLessThan(8 * 60);
+    expect(preview.timeMaxSeconds).toBeLessThan(22 * 60);
+  });
+
+  it('stays short for small crawl limits', () => {
+    const preview = buildPipelineRunPreview({
+      presetId: 'full-audit',
+      configState: {
+        ...buildInitialPipelineConfigState(),
+        max_pages: '30',
+        lighthouse_max_pages: '2',
+      },
+    });
+    expect(preview.timeMaxSeconds).toBeLessThan(6 * 60);
+  });
 });
diff --git a/web/src/lib/pipelineRunPreview.ts b/web/src/lib/pipelineRunPreview.ts
index 52c6bc68..9faa2c32 100644
--- a/web/src/lib/pipelineRunPreview.ts
+++ b/web/src/lib/pipelineRunPreview.ts
@@ -1,19 +1,38 @@
 import { applyPreset, getPresetById, type PipelinePresetId } from '@/components/pipeline/pipelinePresets';
-import { getCrawlPresetById, isCrawlPresetId, type CrawlPresetId } from '@/lib/crawlPresets';
+import { applyCrawlPreset, getCrawlPresetById, isCrawlPresetId, type CrawlPresetId } from '@/lib/crawlPresets';
 import type { PipelineConfigState } from '@/types/api';
 
+type EstimateTier = 'typical' | 'limit';
+
 export interface PipelineRunPhase {
   id: string;
   label: string;
   detail?: string;
 }
 
+export type PipelineTimingPhase =
+  | 'config'
+  | 'crawl'
+  | 'lighthouse'
+  | 'report'
+  | 'keywords'
+  | 'plot'
+  | 'optional';
+
+export interface PipelinePhaseTiming {
+  phase: PipelineTimingPhase;
+  typicalSeconds: number;
+  limitSeconds: number;
+}
+
 export interface PipelineRunPreview {
   phases: PipelineRunPhase[];
   maxCrawlPages: number | null;
+  typicalCrawlPages: number | null;
   lighthousePages: number | null;
   timeMinSeconds: number;
   timeMaxSeconds: number;
+  phaseTimings: PipelinePhaseTiming[];
   configRows: { label: string; value: string }[];
   summaryLines: string[];
 }
@@ -105,20 +124,68 @@ function resolveRunPlan(
   };
 }
 
-function estimateCrawlSeconds(state: PipelineConfigState, maxPages: number): { min: number; max: number } {
+/** Pages likely discovered before hitting the configured crawl cap. */
+function typicalDiscoveredPages(maxPages: number): number {
+  if (maxPages <= 75) return maxPages;
+  return Math.min(maxPages, Math.max(40, Math.round(maxPages * 0.06)));
+}
+
+function lighthouseSamplePages(
+  configured: number,
+  crawlPages: number,
+  tier: EstimateTier,
+): number {
+  const capped = Math.min(Math.max(0, configured), crawlPages);
+  if (capped <= 0) return 0;
+  if (tier === 'limit') return capped;
+  if (capped <= 5) return capped;
+  return Math.min(capped, Math.max(2, Math.round(capped * 0.25)));
+}
+
+function estimateCrawlSeconds(
+  state: PipelineConfigState,
+  pages: number,
+  tier: EstimateTier,
+): number {
+  if (pages <= 0) return 0;
   const concurrency = Math.max(1, num(state.concurrency, 8));
   const politeDelay = Math.max(0, num(state.polite_delay, 0.2));
   const renderMode = String(state.crawl_render_mode ?? 'static').toLowerCase();
   const perPage =
-    renderMode === 'javascript' ? 5.5 : renderMode === 'auto' ? 2.8 : 1.1;
-  const batchSeconds = (maxPages / concurrency) * (perPage + politeDelay);
-  return { min: batchSeconds * 0.75, max: batchSeconds * 1.35 };
+    renderMode === 'javascript' ? 4.8 : renderMode === 'auto' ? 2.4 : 1.0;
+  const batchSeconds = (pages / concurrency) * (perPage + politeDelay);
+  const variance = tier === 'typical' ? 0.9 : 1.2;
+  return batchSeconds * variance;
 }
 
-function estimateLighthouseSeconds(pageCount: number, concurrency: number): { min: number; max: number } {
-  const c = Math.max(1, concurrency);
-  const batch = (pageCount / c) * 55;
-  return { min: batch * 0.85, max: batch * 1.25 };
+function estimateLighthouseSeconds(
+  pageCount: number,
+  concurrency: number,
+  tier: EstimateTier,
+): number {
+  if (pageCount <= 0) return 0;
+  const workers = Math.max(1, concurrency);
+  const perPageWall = 38;
+  const startup = 12;
+  const batch = startup + (pageCount / workers) * perPageWall;
+  return batch * (tier === 'typical' ? 0.92 : 1.12);
+}
+
+function estimateReportSeconds(pages: number, tier: EstimateTier): number {
+  if (pages <= 0) return tier === 'typical' ? 20 : 35;
+  const base = tier === 'typical' ? 25 : 55;
+  const perPage = tier === 'typical' ? 0.012 : 0.035;
+  return base + pages * perPage;
+}
+
+function resolvePreviewState(
+  configState: PipelineConfigState,
+  crawlPresetId: CrawlPresetId | '',
+): PipelineConfigState {
+  if (crawlPresetId && isCrawlPresetId(crawlPresetId)) {
+    return applyCrawlPreset(crawlPresetId, configState);
+  }
+  return configState;
 }
 
 export function buildPipelineRunPreview({
@@ -132,13 +199,20 @@ export function buildPipelineRunPreview({
   customCommand?: string;
   crawlPresetId?: CrawlPresetId | '';
 }): PipelineRunPreview {
-  const plan = resolveRunPlan(presetId, configState, customCommand);
+  const mergedConfig = resolvePreviewState(configState, crawlPresetId);
+  const plan = resolveRunPlan(presetId, mergedConfig, customCommand);
   const { state } = plan;
 
   const maxCrawlPages = plan.includesCrawl ? num(state.max_pages, 500) : null;
-  const lhOnPages = plan.includesLighthouseOnPages
+  const typicalCrawlPages =
+    maxCrawlPages != null ? typicalDiscoveredPages(maxCrawlPages) : null;
+  const lhConfigured = plan.includesLighthouseOnPages
     ? num(state.lighthouse_max_pages, 2)
     : null;
+  const lhOnPages =
+    lhConfigured != null && maxCrawlPages != null
+      ? Math.min(lhConfigured, maxCrawlPages)
+      : lhConfigured;
   const lhStandalone = plan.includesStandaloneLighthouse ? 1 : null;
   const lighthousePages = lhOnPages ?? lhStandalone;
 
@@ -179,28 +253,30 @@ export function buildPipelineRunPreview({
 
   let timeMin = 0;
   let timeMax = 0;
+  const lhConcurrency = num(state.lighthouse_concurrency, 2);
 
-  if (plan.includesCrawl && maxCrawlPages != null) {
-    const crawl = estimateCrawlSeconds(state, maxCrawlPages);
-    timeMin += crawl.min;
-    timeMax += crawl.max;
+  if (plan.includesCrawl && maxCrawlPages != null && typicalCrawlPages != null) {
+    timeMin += estimateCrawlSeconds(state, typicalCrawlPages, 'typical');
+    timeMax += estimateCrawlSeconds(state, maxCrawlPages, 'limit');
   }
   if (plan.includesReport) {
-    const pages = maxCrawlPages ?? num(state.analysis_dup_max_pages, 2000);
-    timeMin += 30 + pages * 0.02;
-    timeMax += 90 + pages * 0.06;
+    const typicalPages = typicalCrawlPages ?? num(state.analysis_dup_max_pages, 2000);
+    const limitPages = maxCrawlPages ?? num(state.analysis_dup_max_pages, 2000);
+    timeMin += estimateReportSeconds(typicalPages, 'typical');
+    timeMax += estimateReportSeconds(limitPages, 'limit');
   }
   if (plan.includesPlot) {
-    timeMin += 20;
-    timeMax += 60;
+    timeMin += 15;
+    timeMax += 45;
   }
-  if (lighthousePages != null && lighthousePages > 0) {
-    const lh = estimateLighthouseSeconds(
-      lighthousePages,
-      num(state.lighthouse_concurrency, 2),
-    );
-    timeMin += lh.min;
-    timeMax += lh.max;
+  if (lhOnPages != null && lhOnPages > 0 && typicalCrawlPages != null && maxCrawlPages != null) {
+    const lhTypical = lighthouseSamplePages(lhOnPages, typicalCrawlPages, 'typical');
+    const lhLimit = lighthouseSamplePages(lhOnPages, maxCrawlPages, 'limit');
+    timeMin += estimateLighthouseSeconds(lhTypical, lhConcurrency, 'typical');
+    timeMax += estimateLighthouseSeconds(lhLimit, lhConcurrency, 'limit');
+  } else if (plan.includesStandaloneLighthouse) {
+    timeMin += estimateLighthouseSeconds(1, 1, 'typical');
+    timeMax += estimateLighthouseSeconds(1, 1, 'limit');
   }
   if (plan.includesGoogle) {
     timeMin += 120;
@@ -214,6 +290,65 @@ export function buildPipelineRunPreview({
   timeMin = Math.max(15, Math.round(timeMin));
   timeMax = Math.max(timeMin + 10, Math.round(timeMax));
 
+  const phaseTimings: PipelinePhaseTiming[] = [];
+  const hasMultiStep =
+    plan.includesCrawl ||
+    plan.includesReport ||
+    plan.includesPlot ||
+    plan.includesLighthouseOnPages ||
+    plan.includesStandaloneLighthouse ||
+    plan.includesGoogle ||
+    plan.includesKeywords;
+  if (hasMultiStep && !plan.command) {
+    phaseTimings.push({ phase: 'config', typicalSeconds: 3, limitSeconds: 8 });
+  }
+  if (plan.includesCrawl && maxCrawlPages != null && typicalCrawlPages != null) {
+    phaseTimings.push({
+      phase: 'crawl',
+      typicalSeconds: estimateCrawlSeconds(state, typicalCrawlPages, 'typical'),
+      limitSeconds: estimateCrawlSeconds(state, maxCrawlPages, 'limit'),
+    });
+  }
+  if (lhOnPages != null && lhOnPages > 0 && typicalCrawlPages != null && maxCrawlPages != null) {
+    phaseTimings.push({
+      phase: 'lighthouse',
+      typicalSeconds: estimateLighthouseSeconds(
+        lighthouseSamplePages(lhOnPages, typicalCrawlPages, 'typical'),
+        lhConcurrency,
+        'typical',
+      ),
+      limitSeconds: estimateLighthouseSeconds(
+        lighthouseSamplePages(lhOnPages, maxCrawlPages, 'limit'),
+        lhConcurrency,
+        'limit',
+      ),
+    });
+  } else if (plan.includesStandaloneLighthouse) {
+    phaseTimings.push({
+      phase: 'lighthouse',
+      typicalSeconds: estimateLighthouseSeconds(1, 1, 'typical'),
+      limitSeconds: estimateLighthouseSeconds(1, 1, 'limit'),
+    });
+  }
+  if (plan.includesReport) {
+    const typicalPages = typicalCrawlPages ?? num(state.analysis_dup_max_pages, 2000);
+    const limitPages = maxCrawlPages ?? num(state.analysis_dup_max_pages, 2000);
+    phaseTimings.push({
+      phase: 'report',
+      typicalSeconds: estimateReportSeconds(typicalPages, 'typical'),
+      limitSeconds: estimateReportSeconds(limitPages, 'limit'),
+    });
+  }
+  if (plan.includesKeywords) {
+    phaseTimings.push({ phase: 'keywords', typicalSeconds: 180, limitSeconds: 720 });
+  }
+  if (plan.includesPlot) {
+    phaseTimings.push({ phase: 'plot', typicalSeconds: 15, limitSeconds: 45 });
+  }
+  if (plan.includesGoogle) {
+    phaseTimings.push({ phase: 'keywords', typicalSeconds: 120, limitSeconds: 360 });
+  }
+
   const renderMode = String(state.crawl_render_mode ?? 'static');
   const crawlPresetLabel =
     crawlPresetId && isCrawlPresetId(crawlPresetId)
@@ -278,9 +413,11 @@ export function buildPipelineRunPreview({
   return {
     phases,
     maxCrawlPages,
+    typicalCrawlPages,
     lighthousePages,
     timeMinSeconds: timeMin,
     timeMaxSeconds: timeMax,
+    phaseTimings,
     configRows,
     summaryLines,
   };
diff --git a/web/src/lib/reportCompare.ts b/web/src/lib/reportCompare.ts
index 38c8f869..f397abcb 100644
--- a/web/src/lib/reportCompare.ts
+++ b/web/src/lib/reportCompare.ts
@@ -50,6 +50,13 @@ export interface LinkStatusChangeRow {
   baselineStatus: string;
 }
 
+export interface UrlMetadataChangeRow {
+  url: string;
+  field: string;
+  baseline: string;
+  current: string;
+}
+
 export interface SeoHealthDeltaRow {
   id: string;
   label: string;
@@ -66,6 +73,7 @@ export interface ReportCompareSummary {
   categoryScores: CategoryScoreRow[];
   seoHealth: SeoHealthDeltaRow[];
   statusChanges: LinkStatusChangeRow[];
+  urlMetadataChanges: UrlMetadataChangeRow[];
   currentGeneratedAt: string | null;
   baselineGeneratedAt: string | null;
   extras: ReportCompareExtras;
@@ -135,6 +143,36 @@ function buildLinkStatusChanges(current: ReportPayload, baseline: ReportPayload)
   return out;
 }
 
+const METADATA_FIELDS: { key: string; label: string }[] = [
+  { key: 'title', label: 'Title' },
+  { key: 'status', label: 'Status' },
+  { key: 'canonical_url', label: 'Canonical' },
+];
+
+function buildUrlMetadataChanges(current: ReportPayload, baseline: ReportPayload): UrlMetadataChangeRow[] {
+  const curMap = new Map<string, ReportLink>();
+  for (const l of current.links ?? []) {
+    const k = normReportUrl(l.url);
+    if (k) curMap.set(k, l);
+  }
+  const out: UrlMetadataChangeRow[] = [];
+  for (const bl of baseline.links ?? []) {
+    const k = normReportUrl(bl.url);
+    if (!k) continue;
+    const cl = curMap.get(k);
+    if (!cl) continue;
+    for (const { key, label } of METADATA_FIELDS) {
+      const curVal = String((cl as unknown as Record<string, unknown>)[key] ?? '').trim();
+      const baseVal = String((bl as unknown as Record<string, unknown>)[key] ?? '').trim();
+      if (curVal !== baseVal) {
+        out.push({ url: cl.url || bl.url, field: label, baseline: baseVal, current: curVal });
+      }
+    }
+  }
+  out.sort((a, b) => a.url.localeCompare(b.url) || a.field.localeCompare(b.field));
+  return out;
+}
+
 function categoryKey(cat: ReportCategory): string {
   return String(cat.id || cat.name || '').trim();
 }
@@ -286,6 +324,7 @@ export function buildReportCompareSummary(
     categoryScores: buildCategoryScores(current, baseline),
     seoHealth: buildSeoHealthDeltas(current, baseline),
     statusChanges: buildLinkStatusChanges(current, baseline),
+    urlMetadataChanges: buildUrlMetadataChanges(current, baseline),
     currentGeneratedAt: current.report_generated_at ?? null,
     baselineGeneratedAt: baseline.report_generated_at ?? null,
     extras: buildReportCompareExtras(current, baseline, extrasLabels),
diff --git a/web/src/lib/siteStructureTree.ts b/web/src/lib/siteStructureTree.ts
index c18dda6a..7c22e3d8 100644
--- a/web/src/lib/siteStructureTree.ts
+++ b/web/src/lib/siteStructureTree.ts
@@ -75,6 +75,14 @@ export function normalizePathname(url: string, expectedHost = ''): string | null
   return p || '/';
 }
 
+/** True when link URL belongs under a path-tree prefix key. */
+export function linkMatchesPathKey(url: string, pathKey: string, expectedHost = ''): boolean {
+  const pathname = normalizePathname(url, expectedHost);
+  if (!pathname) return false;
+  if (!pathKey || pathKey === '/') return true;
+  return pathname === pathKey || pathname.startsWith(`${pathKey}/`);
+}
+
 /** Prefix paths from pathname: `/`, `/blog`, `/blog/a` for `/blog/a`. */
 export function prefixKeysForPathname(pathname: string): string[] {
   if (!pathname || pathname === '/') return ['/'];
diff --git a/web/src/routes.ts b/web/src/routes.ts
index 17676208..591a2a74 100644
--- a/web/src/routes.ts
+++ b/web/src/routes.ts
@@ -17,6 +17,8 @@ export type ViewId =
   | 'gallery'
   | 'search-performance'
   | 'indexation'
+  | 'subdomains'
+  | 'contacts'
   | 'backlinks'
   | 'traffic'
   | 'keywords-explorer'
@@ -42,6 +44,8 @@ const VIEW_IDS = new Set<string>([
   'gallery',
   'search-performance',
   'indexation',
+  'subdomains',
+  'contacts',
   'backlinks',
   'traffic',
   'keywords-explorer',
diff --git a/web/src/server/aiFixSuggestionRoute.test.ts b/web/src/server/aiFixSuggestionRoute.test.ts
new file mode 100644
index 00000000..fecbdfdc
--- /dev/null
+++ b/web/src/server/aiFixSuggestionRoute.test.ts
@@ -0,0 +1,51 @@
+import { describe, expect, it, vi, beforeEach } from 'vitest';
+import { localRequest } from '@/server/testHelpers/routeTestUtils';
+import { makeSpawnChild } from '@/server/testHelpers/routeTestUtils';
+
+const spawnMock = vi.fn();
+vi.mock('child_process', () => ({
+  spawn: (...args: unknown[]) => spawnMock(...args),
+}));
+
+describe('ai/fix-suggestion route', () => {
+  beforeEach(() => {
+    spawnMock.mockReset();
+    vi.resetModules();
+  });
+
+  it('returns 400 when message missing', async () => {
+    const { POST } = await import('../../app/api/ai/fix-suggestion/route');
+    const res = await POST(
+      localRequest('/api/ai/fix-suggestion', {
+        method: 'POST',
+        body: JSON.stringify({ source: 'lighthouse' }),
+      }),
+    );
+    expect(res.status).toBe(400);
+  });
+
+  it('returns suggestion when Python succeeds', async () => {
+    spawnMock.mockImplementation(() =>
+      makeSpawnChild(
+        JSON.stringify({ ok: true, fix: { fix: 'Add preload hint', effort: 'low' } }) + '\n',
+        0,
+      ),
+    );
+    const { POST } = await import('../../app/api/ai/fix-suggestion/route');
+    const res = await POST(
+      localRequest('/api/ai/fix-suggestion', {
+        method: 'POST',
+        body: JSON.stringify({
+          source: 'lighthouse',
+          message: 'Slow LCP',
+          url: 'https://example.com/',
+          context: { kind: 'audit', id: 'lcp' },
+        }),
+      }),
+    );
+    expect(res.status).toBe(200);
+    const body = await res.json();
+    expect(body.ok).toBe(true);
+    expect(body.fix.fix).toMatch(/preload/i);
+  });
+});
diff --git a/web/src/server/exportSitemapRoute.test.ts b/web/src/server/exportSitemapRoute.test.ts
new file mode 100644
index 00000000..a32e54c2
--- /dev/null
+++ b/web/src/server/exportSitemapRoute.test.ts
@@ -0,0 +1,39 @@
+import { describe, expect, it, vi, beforeEach } from 'vitest';
+import { localRequest } from '@/server/testHelpers/routeTestUtils';
+
+const spawnMock = vi.fn();
+vi.mock('child_process', () => ({
+  spawn: (...args: unknown[]) => spawnMock(...args),
+}));
+
+describe('report/export-sitemap route', () => {
+  beforeEach(() => {
+    spawnMock.mockReset();
+    vi.resetModules();
+  });
+
+  it('returns xml on success', async () => {
+    spawnMock.mockImplementation(() => ({
+      stdout: { on: (_: string, cb: (c: Buffer) => void) => cb(Buffer.from('<urlset></urlset>')) },
+      stderr: { on: () => undefined },
+      on: (_: string, cb: (code: number) => void) => cb(0),
+    }));
+    const { GET } = await import('../../app/api/report/export-sitemap/route');
+    const res = await GET(localRequest('/api/report/export-sitemap?reportId=12'));
+    expect(res.status).toBe(200);
+    expect(res.headers.get('Content-Type')).toContain('application/xml');
+    const text = await res.text();
+    expect(text).toContain('urlset');
+  });
+
+  it('returns 500 when python fails', async () => {
+    spawnMock.mockImplementation(() => ({
+      stdout: { on: () => undefined },
+      stderr: { on: (_: string, cb: (c: Buffer) => void) => cb(Buffer.from('boom')) },
+      on: (_: string, cb: (code: number) => void) => cb(1),
+    }));
+    const { GET } = await import('../../app/api/report/export-sitemap/route');
+    const res = await GET(localRequest('/api/report/export-sitemap?reportId=12'));
+    expect(res.status).toBe(500);
+  });
+});
diff --git a/web/src/server/filtersRoute.test.ts b/web/src/server/filtersRoute.test.ts
new file mode 100644
index 00000000..10d58b2e
--- /dev/null
+++ b/web/src/server/filtersRoute.test.ts
@@ -0,0 +1,50 @@
+import { describe, expect, it, vi, beforeEach } from 'vitest';
+import { localRequest } from '@/server/testHelpers/routeTestUtils';
+
+const listMock = vi.fn();
+const upsertMock = vi.fn();
+const deleteMock = vi.fn();
+
+vi.mock('@/server/savedFiltersDb', () => ({
+  listSavedFilters: (...args: unknown[]) => listMock(...args),
+  upsertSavedFilter: (...args: unknown[]) => upsertMock(...args),
+  deleteSavedFilter: (...args: unknown[]) => deleteMock(...args),
+}));
+
+describe('filters route', () => {
+  beforeEach(() => {
+    listMock.mockReset();
+    upsertMock.mockReset();
+    deleteMock.mockReset();
+    vi.resetModules();
+  });
+
+  it('GET returns 400 without propertyId', async () => {
+    const { GET } = await import('../../app/api/filters/route');
+    const res = await GET(localRequest('/api/filters'));
+    expect(res.status).toBe(400);
+  });
+
+  it('GET lists saved filters', async () => {
+    listMock.mockResolvedValue([{ name: 'Broken', filterJson: { status: '404' } }]);
+    const { GET } = await import('../../app/api/filters/route');
+    const res = await GET(localRequest('/api/filters?propertyId=7'));
+    expect(res.status).toBe(200);
+    const body = await res.json();
+    expect(body.filters).toHaveLength(1);
+    expect(listMock).toHaveBeenCalledWith(7);
+  });
+
+  it('POST upserts a filter', async () => {
+    upsertMock.mockResolvedValue(undefined);
+    const { POST } = await import('../../app/api/filters/route');
+    const res = await POST(
+      localRequest('/api/filters', {
+        method: 'POST',
+        body: JSON.stringify({ propertyId: 3, name: 'Deep pages', filterJson: { minDepth: '3' } }),
+      }),
+    );
+    expect(res.status).toBe(200);
+    expect(upsertMock).toHaveBeenCalledWith(3, 'Deep pages', { minDepth: '3' });
+  });
+});
diff --git a/web/src/server/issuesFixSuggestionRoute.test.ts b/web/src/server/issuesFixSuggestionRoute.test.ts
index 1426b5d6..037d1194 100644
--- a/web/src/server/issuesFixSuggestionRoute.test.ts
+++ b/web/src/server/issuesFixSuggestionRoute.test.ts
@@ -26,7 +26,10 @@ describe('issues/fix-suggestion route', () => {
 
   it('returns suggestion when Python succeeds', async () => {
     spawnMock.mockImplementation(() =>
-      makeSpawnChild(JSON.stringify({ ok: true, suggestion: 'Add preload hint' }) + '\n', 0),
+      makeSpawnChild(
+        JSON.stringify({ ok: true, fix: { fix: 'Add preload hint', effort: 'low' } }) + '\n',
+        0,
+      ),
     );
     const { POST } = await import('../../app/api/issues/fix-suggestion/route');
     const res = await POST(
@@ -37,6 +40,7 @@ describe('issues/fix-suggestion route', () => {
     );
     expect(res.status).toBe(200);
     const body = await res.json();
-    expect(body.suggestion).toMatch(/preload/i);
+    expect(body.ok).toBe(true);
+    expect(body.fix.fix).toMatch(/preload/i);
   });
 });
diff --git a/web/src/server/keywordsCompetitorImportRoute.test.ts b/web/src/server/keywordsCompetitorImportRoute.test.ts
new file mode 100644
index 00000000..747583f8
--- /dev/null
+++ b/web/src/server/keywordsCompetitorImportRoute.test.ts
@@ -0,0 +1,56 @@
+import { describe, expect, it, vi, beforeEach } from 'vitest';
+import { localRequest } from '@/server/testHelpers/routeTestUtils';
+
+const spawnMock = vi.fn();
+const queryMock = vi.fn();
+
+vi.mock('child_process', () => ({
+  spawn: (...args: unknown[]) => spawnMock(...args),
+}));
+
+vi.mock('@/server/db', () => ({
+  withDb: async (fn: (client: { query: typeof queryMock }) => Promise<void>) => fn({ query: queryMock }),
+}));
+
+describe('keywords/competitor-import route', () => {
+  beforeEach(() => {
+    spawnMock.mockReset();
+    queryMock.mockReset();
+    vi.resetModules();
+  });
+
+  it('returns 400 when csvText missing', async () => {
+    const { POST } = await import('../../app/api/keywords/competitor-import/route');
+    const res = await POST(
+      localRequest('/api/keywords/competitor-import', {
+        method: 'POST',
+        body: JSON.stringify({ propertyId: 1, competitor: 'rival.com', csvText: '  ' }),
+      }),
+    );
+    expect(res.status).toBe(400);
+  });
+
+  it('imports competitor keywords on success', async () => {
+    spawnMock.mockImplementation(() => ({
+      stdout: { on: (_: string, cb: (c: Buffer) => void) => cb(Buffer.from('{"count":1,"rows":[{"keyword":"kw","competitor":"rival.com"}]}')) },
+      stderr: { on: () => undefined },
+      stdin: { write: () => undefined, end: () => undefined },
+      on: (_: string, cb: (code: number) => void) => cb(0),
+    }));
+    queryMock.mockResolvedValue(undefined);
+    const { POST } = await import('../../app/api/keywords/competitor-import/route');
+    const res = await POST(
+      localRequest('/api/keywords/competitor-import', {
+        method: 'POST',
+        body: JSON.stringify({
+          propertyId: 1,
+          competitor: 'rival.com',
+          csvText: 'Keyword,Volume\nkw,100\n',
+        }),
+      }),
+    );
+    expect(res.status).toBe(200);
+    const body = await res.json();
+    expect(body.count).toBe(1);
+  });
+});
diff --git a/web/src/server/savedFiltersDb.ts b/web/src/server/savedFiltersDb.ts
new file mode 100644
index 00000000..924b31db
--- /dev/null
+++ b/web/src/server/savedFiltersDb.ts
@@ -0,0 +1,56 @@
+import { withDb } from '@/server/db';
+
+export interface SavedCrawlFilterRow {
+  id: number;
+  propertyId: number;
+  name: string;
+  filterJson: Record<string, unknown>;
+  createdAt: string;
+}
+
+export async function listSavedFilters(propertyId: number): Promise<SavedCrawlFilterRow[]> {
+  return withDb(async (client) => {
+    const res = await client.query<{
+      id: number;
+      property_id: number;
+      name: string;
+      filter_json: Record<string, unknown>;
+      created_at: string;
+    }>(
+      `SELECT id, property_id, name, filter_json, created_at
+       FROM saved_crawl_filters WHERE property_id = $1 ORDER BY name`,
+      [propertyId],
+    );
+    return res.rows.map((r) => ({
+      id: r.id,
+      propertyId: r.property_id,
+      name: r.name,
+      filterJson: r.filter_json || {},
+      createdAt: r.created_at,
+    }));
+  });
+}
+
+export async function upsertSavedFilter(
+  propertyId: number,
+  name: string,
+  filterJson: Record<string, unknown>,
+): Promise<void> {
+  await withDb(async (client) => {
+    await client.query(
+      `INSERT INTO saved_crawl_filters (property_id, name, filter_json)
+       VALUES ($1, $2, $3::jsonb)
+       ON CONFLICT (property_id, name) DO UPDATE SET filter_json = EXCLUDED.filter_json`,
+      [propertyId, name, JSON.stringify(filterJson || {})],
+    );
+  });
+}
+
+export async function deleteSavedFilter(propertyId: number, name: string): Promise<void> {
+  await withDb(async (client) => {
+    await client.query(`DELETE FROM saved_crawl_filters WHERE property_id = $1 AND name = $2`, [
+      propertyId,
+      name,
+    ]);
+  });
+}
diff --git a/web/src/strings.json b/web/src/strings.json
index 5c51dfdb..078409c6 100644
--- a/web/src/strings.json
+++ b/web/src/strings.json
@@ -107,6 +107,14 @@
       "label": "Indexation",
       "section": "Search & analytics"
     },
+    "subdomains": {
+      "label": "Subdomains",
+      "section": "Property"
+    },
+    "contacts": {
+      "label": "Contact signals",
+      "section": "Property"
+    },
     "backlinks": {
       "label": "Backlinks",
       "section": "Search & analytics"
@@ -163,12 +171,13 @@
       "title": "Run preview",
       "hint": "Estimated scope from your audit type and crawl settings.",
       "estimatedTime": "Est. duration",
+      "liveEstimatedTime": "Live estimate",
       "maxPagesLabel": "Max pages",
       "lighthousePagesLabel": "Lighthouse URLs",
       "stepsLabel": "Pipeline steps",
       "whatRunsLabel": "What will run",
       "configPreviewLabel": "Configuration preview",
-      "estimateDisclaimer": "Duration is an estimate — actual time depends on site speed, server response, and network conditions."
+      "estimateDisclaimer": "Duration is an estimate for a typical site size up to your crawl limit — most sites finish faster than the upper bound."
     },
     "wizardAdvancedSettings": "Advanced settings",
     "wizardUrlHint": "Enter the site you want to analyze.",
@@ -1146,6 +1155,7 @@
       "title": "All URLs",
       "tabs": {
         "urls": "All URLs",
+        "anchors": "Inlink anchor text",
         "charts": "Charts"
       },
       "chartsEmpty": "No distribution charts are available for this crawl yet.",
@@ -1153,6 +1163,28 @@
       "showingResults": "Analyze discovered URLs. Showing",
       "resultsSuffix": "results.",
       "explorerHint": "Each row is a crawled URL. \"Inlinks\" is how many internal links point to this page in this crawl. Sort columns or open URL details for the full breakdown.",
+      "siteTechnicalIssuesTitle": "Site-wide broken links & redirects",
+      "siteTechnicalIssuesHint": "Crawl-level HTTP issues with AI fix suggestions. Open a URL in the inspector for page-specific detail.",
+      "linkAttributesTitle": "Link attributes",
+      "linkAttrTotal": "Total links",
+      "linkAttrInternal": "Internal",
+      "linkAttrNofollow": "Nofollow internal",
+      "linkAttrSponsored": "Sponsored internal",
+      "linkAttrExternal": "External",
+      "linkAttrFollow": "Follow internal",
+      "linkAttrUgc": "UGC internal",
+      "chartLinkScopeTitle": "Internal vs external links",
+      "chartLinkScopeHint": "All link edges discovered in this crawl",
+      "chartInternalAttrsTitle": "Internal link attributes",
+      "chartInternalAttrsHint": "Follow, nofollow, sponsored, and UGC on internal links",
+      "chartTopAnchorsTitle": "Top anchor text",
+      "chartTopAnchorsHint": "Anchor text receiving the most internal inlinks",
+      "chartTopTargetsTitle": "Top linked pages",
+      "chartTopTargetsHint": "Pages receiving the most internal inlinks",
+      "inlinkAnchorMatrix": "Inlink anchor text",
+      "inlinkTarget": "Target URL",
+      "inlinkAnchor": "Anchor text",
+      "inlinkCount": "Inlinks",
       "noTitle": "No title",
       "filterAllPages": "All pages",
       "filterOrphans": "Orphans (0 inlinks)",
@@ -1341,6 +1373,52 @@
       "gapsHint": "URLs present in one source but missing from another. Measured from crawl; Search Console from Google API.",
       "sitemapOnlyList": "In sitemap but not crawled"
     },
+    "subdomains": {
+      "title": "Subdomains",
+      "subtitle": "Passive host inventory from crawl, Search Console, and certificate transparency.",
+      "disabledHint": "Subdomain discovery is disabled in pipeline settings. Enable it and re-run the audit.",
+      "emptyHint": "No subdomain data in this report. Re-run the audit to build the inventory.",
+      "ctWarning": "Certificate transparency lookup failed; crawl and GSC sources are still shown.",
+      "apex": "Apex domain",
+      "inScopeHosts": "In-scope hosts",
+      "gscNotCrawled": "GSC hosts not crawled",
+      "outOfScope": "Out of scope",
+      "gscGapTitle": "Search Console hosts missing from crawl",
+      "gscGapHint": "These hosts appear in GSC but had no URLs in this crawl — often separate subdomains or out-of-scope properties.",
+      "moreHosts": "+{count} more hosts",
+      "viewIndexation": "View indexation gaps",
+      "hostsTitle": "Discovered hosts (in scope)",
+      "colHost": "Host",
+      "colSources": "Sources",
+      "colCrawl": "In crawl",
+      "colGsc": "In GSC",
+      "colCrawlUrls": "Crawl URLs",
+      "colGscUrls": "GSC URLs",
+      "noSearchResults": "No hosts match your search.",
+      "outOfScopeTitle": "Out-of-scope hosts",
+      "outOfScopeHint": "Discovered via CT or external links; not counted as part of this property."
+    },
+    "contacts": {
+      "title": "Contact signals",
+      "subtitle": "Business contact data sourced from crawl, structured data, security.txt, and RDAP.",
+      "emptyHint": "No contact signals in this report. Re-run the audit after crawling contact pages.",
+      "notesTitle": "Consistency notes",
+      "primaryPageTitle": "Primary contact page",
+      "noSearchResults": "No contact signals match your search.",
+      "emailsTitle": "Email addresses",
+      "phonesTitle": "Phone numbers",
+      "addressesTitle": "Addresses",
+      "organizationsTitle": "Organization names",
+      "noEmails": "No email addresses detected.",
+      "noPhones": "No phone numbers detected.",
+      "noAddresses": "No postal addresses detected.",
+      "noOrganizations": "No organization names in structured data.",
+      "colValue": "Value",
+      "colSources": "Sources",
+      "colPages": "Found on",
+      "morePages": "+{count} more pages",
+      "provenanceHint": "Each value lists where it was found. No third-party email enrichment is used."
+    },
     "searchPerformance": {
       "title": "Search Performance",
       "subtitle": "Google Search Console queries, pages, and crawl alignment for {range}",
@@ -1768,6 +1846,16 @@
         "provenance": "Source",
         "close": "Close"
       },
+      "competitorGap": {
+        "title": "Imported competitor keywords",
+        "hint": "From Ahrefs or Semrush CSV import. Re-run the audit after import to persist in the report payload.",
+        "empty": "No competitor keywords in this report yet. Import a CSV above, then run Keywords or a full audit.",
+        "colKeyword": "Keyword",
+        "colCompetitor": "Competitor",
+        "colVolume": "Volume",
+        "colPosition": "Position",
+        "colUrl": "Landing URL"
+      },
       "cannib": {
         "empty": "No cannibalisation detected. Each keyword appears to rank from a unique page.",
         "emptyFiltered": "No cannibalisation issues match your search.",
@@ -1934,7 +2022,9 @@
       "print": "Print",
       "downloadPdf": "Download PDF",
       "downloadCsv": "Download CSV",
-      "downloadJson": "Download JSON"
+      "downloadJson": "Download JSON",
+      "downloadWorkbook": "Download crawl workbook (ZIP)",
+      "downloadSitemap": "Download XML sitemap"
     },
     "overview": {
       "openExportPage": "Export report",
@@ -2051,6 +2141,11 @@
       "lhCategoryScores": "Lighthouse category scores",
       "lhCategoryHint": "From the attached Lighthouse summary (typically the audited landing or representative URL). Scores are 0–100; bars use the same good / needs work / poor coloring as elsewhere.",
       "healthByCategory": "Health by Category",
+      "portfolioBenchmarkTitle": "Portfolio benchmark",
+      "portfolioBenchmarkHint": "Compare this property’s average category score to the median across all properties on this instance.",
+      "portfolioPropertyScore": "This property",
+      "portfolioMedianScore": "Portfolio median",
+      "portfolioDelta": "Vs median",
       "noCategorySearch": "No categories match your search.",
       "noCategoryData": "No category data.",
       "scoreGood": "Good",
@@ -2062,6 +2157,13 @@
       "sitemapXml": "sitemap.xml:",
       "sitemapValid": " (valid)",
       "sitemapInvalid": " (invalid or unparsed)",
+      "adsTxt": "ads.txt:",
+      "adsTxtValid": " (valid)",
+      "adsTxtInvalid": " (present, invalid)",
+      "securityTxt": "security.txt:",
+      "securityTxtContacts": " ({count} contact(s))",
+      "viewSubdomains": "View subdomains",
+      "viewContacts": "View contact signals",
       "recommendations": "Recommendations",
       "noRecSearch": "No recommendations match your search.",
       "topPagesTitle": "Most important pages",
@@ -2152,6 +2254,15 @@
       "twitterTags": "twitter card tags",
       "thinPages": "Thin Pages",
       "under300": "under 300 words",
+      "richResultsTitle": "Structured data (Rich Results)",
+      "richResultsHint": "Sample of 2xx URLs checked via GSC URL Inspection, Rich Results API (when keyed), or crawl JSON-LD heuristics.",
+      "richResultsMeta": "{gsc} GSC · {api} API · {heuristic} estimated from crawl",
+      "richResultsUpgradeHint": "Connect Search Console or add a Rich Results API key for authoritative validation.",
+      "richResultsEmpty": "No Rich Results validation rows.",
+      "richResultsColUrl": "URL",
+      "richResultsColStatus": "Status",
+      "richResultsColSource": "Source",
+      "richResultsColMessage": "Details",
       "i18nTitle": "Internationalization (crawl)",
       "pages2xx": "2xx pages",
       "missingHtmlLang": "Missing <html lang>",
@@ -2463,6 +2574,20 @@
       "brokenMessage": "Broken / error response ({status})",
       "redirectMessage": "Redirect {status}"
     },
+    "pageImprove": {
+      "coachButton": "Get AI coach",
+      "coachLoading": "Generating coach…",
+      "coachFailed": "Page coach failed",
+      "coachTitle": "AI recommendations"
+    },
+    "aiSuggestion": {
+      "button": "AI fix suggestion",
+      "regenerate": "Regenerate AI fix",
+      "loading": "Generating…",
+      "failed": "AI fix unavailable",
+      "empty": "No suggestion returned.",
+      "label": "AI insights"
+    },
     "copyBtn": {
       "copy": "Copy",
       "copied": "Copied!",
diff --git a/web/src/types/fixSuggestion.ts b/web/src/types/fixSuggestion.ts
new file mode 100644
index 00000000..b2e22543
--- /dev/null
+++ b/web/src/types/fixSuggestion.ts
@@ -0,0 +1,35 @@
+export type FixSuggestionSource =
+  | 'issue'
+  | 'lighthouse'
+  | 'security'
+  | 'browser'
+  | 'seo_content'
+  | 'technical';
+
+export interface FixSuggestionRequest {
+  source: FixSuggestionSource;
+  message: string;
+  url?: string;
+  refresh?: boolean;
+  context?: Record<string, unknown>;
+}
+
+export interface FixSuggestionFix {
+  fix: string;
+  effort?: 'low' | 'medium' | 'high';
+}
+
+export interface FixSuggestionResponse {
+  ok: boolean;
+  fix?: FixSuggestionFix | string;
+  cached?: boolean;
+  provenance?: string;
+  error?: string;
+}
+
+export function extractFixText(payload: FixSuggestionResponse): string {
+  const raw = payload.fix;
+  if (typeof raw === 'string') return raw.trim();
+  if (raw && typeof raw === 'object' && typeof raw.fix === 'string') return raw.fix.trim();
+  return '';
+}
diff --git a/web/src/types/index.ts b/web/src/types/index.ts
index d0f5f718..25625c44 100644
--- a/web/src/types/index.ts
+++ b/web/src/types/index.ts
@@ -41,6 +41,10 @@ export type {
   DepthDistribution,
   TechStackSummary,
   SiteLevelChecks,
+  ContactIntelligence,
+  ContactIntelligenceEntry,
+  SubdomainInventory,
+  SubdomainHostEntry,
   NerSiteSummary,
   LanguageSummary,
   KeywordOpportunities,
@@ -77,6 +81,9 @@ export type {
   ContentAnalyticsStats,
   SimilarInternalRow,
   TechStackEntry,
+  PortfolioBenchmark,
+  RichResultsValidationRow,
+  CompetitorKeywordGapRow,
 } from '@/types/report';
 
 export type {
diff --git a/web/src/types/report.ts b/web/src/types/report.ts
index 0ee044b4..eacce712 100644
--- a/web/src/types/report.ts
+++ b/web/src/types/report.ts
@@ -43,6 +43,10 @@ export interface ReportIssue {
   recommendation?: string;
   llm_recommendation?: string;
   llm_fix_effort?: string;
+  impact_score?: number;
+  gsc_clicks?: number;
+  gsc_impressions?: number;
+  ga4_sessions?: number;
   type?: string;
   severity?: string;
   finding_type?: string;
@@ -330,6 +334,49 @@ export interface SiteLevelChecks {
   robots_present?: boolean;
   sitemap_present?: boolean;
   sitemap_valid?: boolean;
+  ads_txt_present?: boolean;
+  ads_txt_valid?: boolean;
+  ads_txt_line_count?: number;
+  ads_txt_issues?: string[];
+  security_txt_present?: boolean;
+  security_txt_valid?: boolean;
+  security_txt_contact?: string[];
+  security_txt_expires?: string | null;
+}
+
+export interface ContactIntelligenceEntry {
+  value?: string;
+  sources?: string[];
+  urls?: string[];
+}
+
+export interface ContactIntelligence {
+  emails?: ContactIntelligenceEntry[];
+  phones?: ContactIntelligenceEntry[];
+  addresses?: ContactIntelligenceEntry[];
+  organization_names?: ContactIntelligenceEntry[];
+  primary_contact_page?: string | null;
+  consistency_notes?: string[];
+}
+
+export interface SubdomainHostEntry {
+  host?: string;
+  sources?: string[];
+  in_crawl?: boolean;
+  in_gsc?: boolean;
+  url_count_crawl?: number;
+  url_count_gsc?: number;
+  in_scope?: boolean;
+}
+
+export interface SubdomainInventory {
+  apex?: string;
+  hosts?: SubdomainHostEntry[];
+  gsc_hosts_not_crawled?: string[];
+  out_of_scope_discovered?: string[];
+  fetched_at?: string;
+  disabled?: boolean;
+  crtsh_error?: string;
 }
 
 export interface NerSiteSummary {
@@ -357,6 +404,32 @@ export interface KeywordOpportunities {
   token_topic_clusters?: TopicCluster[];
 }
 
+export interface LinkEdgeRow {
+  from_url?: string;
+  to_url?: string;
+  anchor_text?: string;
+  rel?: string;
+  is_nofollow?: boolean;
+  is_sponsored?: boolean;
+  is_ugc?: boolean;
+  link_type?: string;
+}
+
+export interface LinkRelSummary {
+  total_edges?: number;
+  internal_edges?: number;
+  nofollow_internal?: number;
+  sponsored_internal?: number;
+  ugc_internal?: number;
+  external_edges?: number;
+}
+
+export interface InlinkAnchorRow {
+  target_url?: string;
+  anchor_text?: string;
+  inlink_count?: number;
+}
+
 export interface GraphEdge {
   from?: string;
   to?: string;
@@ -422,6 +495,43 @@ export interface BingBacklinksSummary {
   error?: string;
 }
 
+export type PortfolioBenchmarkStatus = 'ok' | 'single_property' | 'unavailable' | 'error';
+
+export interface PortfolioBenchmark {
+  median_health_score?: number | null;
+  property_health_score?: number | null;
+  property_count?: number;
+  status?: PortfolioBenchmarkStatus;
+  message?: string;
+}
+
+export interface RichResultsValidationRow {
+  url: string;
+  status?: string;
+  provenance?: string;
+  source?: string;
+  message?: string;
+  verdict?: string;
+  schema_types?: string[];
+  issues?: string[];
+}
+
+export interface RichResultsMeta {
+  checked?: number;
+  gsc_count?: number;
+  api_count?: number;
+  heuristic_count?: number;
+}
+
+export interface CompetitorKeywordGapRow {
+  keyword: string;
+  competitor?: string;
+  volume?: string | number | null;
+  position?: string | number | null;
+  url?: string | null;
+  source?: string;
+}
+
 export interface ReportPayload {
   site_name?: string;
   crawl_run_id?: number;
@@ -481,11 +591,20 @@ export interface ReportPayload {
   depth_distribution?: DepthDistribution;
   tech_stack_summary?: TechStackSummary;
   site_level?: SiteLevelChecks;
+  contact_intelligence?: ContactIntelligence;
+  subdomains?: SubdomainInventory;
   ner_site_summary?: NerSiteSummary;
   language_summary?: LanguageSummary;
   keyword_opportunities?: KeywordOpportunities;
   graph_nodes?: GraphNode[];
   graph_edges?: GraphEdge[];
+  link_edges?: LinkEdgeRow[];
+  link_rel_summary?: LinkRelSummary;
+  inlink_anchor_matrix?: InlinkAnchorRow[];
+  portfolio_benchmark?: PortfolioBenchmark;
+  rich_results_validation?: RichResultsValidationRow[];
+  rich_results_meta?: RichResultsMeta;
+  competitor_keyword_gap?: CompetitorKeywordGapRow[];
   recommendations?: string[];
   mime_labels?: string[];
   mime_values?: number[];
@@ -532,6 +651,8 @@ export interface ReportLink {
   page_error_count?: number;
   has_browser_errors?: boolean;
   custom_extract?: string;
+  /** JSON string or parsed map from XPath/CSS custom extractors */
+  custom_fields?: string | Record<string, string>;
 }
 
 export interface CrawlSegmentEntry {
diff --git a/web/src/utils/linkExport.ts b/web/src/utils/linkExport.ts
index 915a0707..bd34833d 100644
--- a/web/src/utils/linkExport.ts
+++ b/web/src/utils/linkExport.ts
@@ -1,4 +1,5 @@
 import type { ReportLink } from '@/types';
+import { collectCustomFieldKeys, parseLinkCustomFields } from '@/lib/customFields';
 
 function escapeCsv(v: unknown): string {
   if (v == null) return '';
@@ -9,6 +10,7 @@ function escapeCsv(v: unknown): string {
 export function exportLinksCsv(links: ReportLink[], filename = 'crawl_urls.csv'): void {
   if (!links.length) return;
   const hasCustomExtract = links.some((l) => l.custom_extract);
+  const customFieldKeys = collectCustomFieldKeys(links);
   const cols = [
     'url',
     'status',
@@ -19,14 +21,15 @@ export function exportLinksCsv(links: ReportLink[], filename = 'crawl_urls.csv')
     'word_count',
     'response_time_ms',
     ...(hasCustomExtract ? ['custom_extract'] : []),
+    ...customFieldKeys,
   ];
   const lines = [
     cols.join(','),
-    ...links.map((l) =>
-      cols
-        .map((c) => escapeCsv((l as unknown as Record<string, unknown>)[c]))
-        .join(','),
-    ),
+    ...links.map((l) => {
+      const fields = parseLinkCustomFields(l);
+      const row: Record<string, unknown> = { ...l, ...fields };
+      return cols.map((c) => escapeCsv(row[c])).join(',');
+    }),
   ];
   const blob = new Blob([lines.join('\n')], { type: 'text/csv;charset=utf-8;' });
   const url = URL.createObjectURL(blob);
diff --git a/web/src/views/CompareReports.tsx b/web/src/views/CompareReports.tsx
index a4fd3212..5257949d 100644
--- a/web/src/views/CompareReports.tsx
+++ b/web/src/views/CompareReports.tsx
@@ -2,7 +2,7 @@
 
 import { useMemo, useState, useCallback } from 'react';
 import Link from 'next/link';
-import { useSearchParams } from 'next/navigation';
+import { usePathname, useRouter, useSearchParams } from 'next/navigation';
 import { useUrlTab } from '@/hooks/useUrlTab';
 import {
   ArrowLeftRight,
@@ -36,6 +36,7 @@ import {
   CompareLinksPanel,
   CompareGooglePanel,
 } from '../components/compare/CompareTabPanels';
+import CompareUrlMetadataTable from '../components/compare/CompareUrlMetadataTable';
 import dynamic from 'next/dynamic';
 
 const CompareOverviewCharts = dynamic(
@@ -58,9 +59,9 @@ type CompareTab =
   | 'links'
   | 'google'
   | 'audit';
-type UrlTab = 'all' | 'new' | 'removed' | 'content' | 'structure';
+type UrlTab = 'all' | 'new' | 'removed' | 'content' | 'structure' | 'fields';
 
-const URL_TABS = ['all', 'new', 'removed', 'content', 'structure'] as const;
+const URL_TABS = ['all', 'new', 'removed', 'content', 'structure', 'fields'] as const;
 
 function filterUrls(urls: string[], query: string): string[] {
   const q = query.trim().toLowerCase();
@@ -146,17 +147,27 @@ const TAB_KEYS = [
 ] as const satisfies readonly CompareTab[];
 
 export default function CompareReports({ searchQuery = '' }: ViewProps) {
+  const router = useRouter();
+  const pathname = usePathname();
   const searchParams = useSearchParams();
-  const [tab, setTabRaw] = useUrlTab(TAB_KEYS, 'overview', 'tab');
-  const [urlTab, setUrlTabRaw] = useUrlTab(URL_TABS, 'all', 'urlTab');
+  const [tab] = useUrlTab(TAB_KEYS, 'overview', 'tab');
+  const [urlTab, setUrlTab] = useUrlTab(URL_TABS, 'all', 'urlTab');
   const setTab = useCallback(
     (next: CompareTab) => {
-      setTabRaw(next);
-      if (next !== 'urls') setUrlTabRaw('all');
+      const params = new URLSearchParams(searchParams.toString());
+      if (next === 'overview') {
+        params.delete('tab');
+      } else {
+        params.set('tab', next);
+      }
+      if (next !== 'urls') {
+        params.delete('urlTab');
+      }
+      const q = params.toString();
+      router.replace(q ? `${pathname}?${q}` : pathname, { scroll: false });
     },
-    [setTabRaw, setUrlTabRaw],
+    [router, pathname, searchParams],
   );
-  const setUrlTab = setUrlTabRaw;
   const [copyHint, setCopyHint] = useState('');
   const { reportList, reportCompare, compareReportId, selectedReportId, loading, error } = useReport();
   const vc = strings.views.compare;
@@ -204,6 +215,20 @@ export default function CompareReports({ searchQuery = '' }: ViewProps) {
     return all.sort((a, b) => a.localeCompare(b));
   }, [urlLists, urlTab]);
 
+  const metadataFiltered = useMemo(() => {
+    if (!reportCompare) return [];
+    const q = searchQuery.trim().toLowerCase();
+    const rows = reportCompare.urlMetadataChanges || [];
+    if (!q) return rows;
+    return rows.filter(
+      (r) =>
+        r.url.toLowerCase().includes(q) ||
+        r.field.toLowerCase().includes(q) ||
+        r.baseline.toLowerCase().includes(q) ||
+        r.current.toLowerCase().includes(q),
+    );
+  }, [reportCompare, searchQuery]);
+
   const statusFiltered = useMemo(() => {
     if (!reportCompare) return [];
     const q = searchQuery.trim().toLowerCase();
@@ -274,8 +299,9 @@ export default function CompareReports({ searchQuery = '' }: ViewProps) {
       { id: 'removed', label: vc.urlTabs.removed, count: urlLists.removedUrls.length },
       { id: 'content', label: vc.urlTabs.content, count: urlLists.contentChanged.length },
       { id: 'structure', label: vc.urlTabs.structure, count: urlLists.structureChanged.length },
+      { id: 'fields', label: 'Title / canonical', count: reportCompare?.urlMetadataChanges?.length ?? 0 },
     ];
-  }, [urlLists, vc.urlTabs]);
+  }, [urlLists, vc.urlTabs, reportCompare?.urlMetadataChanges?.length]);
 
   return (
     <PageLayout className="space-y-6">
@@ -463,6 +489,11 @@ export default function CompareReports({ searchQuery = '' }: ViewProps) {
                 ))}
               </div>
               {copyHint ? <p className="text-xs text-emerald-700 dark:text-emerald-400">{copyHint}</p> : null}
+              {urlTab === 'fields' ? (
+                <Card shadow>
+                  <CompareUrlMetadataTable rows={metadataFiltered} emptyLabel={vc.noneInCategory} />
+                </Card>
+              ) : (
               <Card shadow>
                 <UrlDiffTable
                   urls={activeUrlList}
@@ -471,6 +502,7 @@ export default function CompareReports({ searchQuery = '' }: ViewProps) {
                   copyLabel={vc.copyUrls}
                 />
               </Card>
+              )}
             </div>
           ) : null}
 
diff --git a/web/src/views/Contacts.tsx b/web/src/views/Contacts.tsx
new file mode 100644
index 00000000..8041de33
--- /dev/null
+++ b/web/src/views/Contacts.tsx
@@ -0,0 +1,191 @@
+'use client';
+
+import { useMemo } from 'react';
+import { Contact2, ExternalLink } from 'lucide-react';
+import { useReport } from '../context/useReport';
+import { strings, format } from '../lib/strings';
+import {
+  PageLayout,
+  PageHeader,
+  Card,
+  Badge,
+  Table,
+  TableHead,
+  TableHeadCell,
+  TableBody,
+  TableRow,
+  TableCell,
+} from '../components';
+import type { ContactIntelligenceEntry, ViewProps } from '@/types';
+
+type ContactSection = 'emails' | 'phones' | 'addresses' | 'organization_names';
+
+function filterEntries(
+  items: ContactIntelligenceEntry[] | undefined,
+  q: string,
+): ContactIntelligenceEntry[] {
+  const list = items || [];
+  if (!q) return list;
+  return list.filter((row) => {
+    const value = (row.value || '').toLowerCase();
+    const sources = (row.sources || []).join(' ').toLowerCase();
+    const urls = (row.urls || []).join(' ').toLowerCase();
+    return value.includes(q) || sources.includes(q) || urls.includes(q);
+  });
+}
+
+function ContactSectionTable({
+  title,
+  rows,
+  emptyLabel,
+}: {
+  title: string;
+  rows: ContactIntelligenceEntry[];
+  emptyLabel: string;
+}) {
+  const vc = strings.views.contacts;
+  if (rows.length === 0) {
+    return (
+      <Card className="mb-6">
+        <h3 className="text-sm font-semibold text-foreground mb-2">{title}</h3>
+        <p className="text-sm text-muted-foreground">{emptyLabel}</p>
+      </Card>
+    );
+  }
+  return (
+    <Card className="mb-6">
+      <h3 className="text-sm font-semibold text-foreground mb-4">{title}</h3>
+      <div className="overflow-x-auto">
+        <Table>
+          <TableHead>
+            <TableRow>
+              <TableHeadCell>{vc.colValue}</TableHeadCell>
+              <TableHeadCell>{vc.colSources}</TableHeadCell>
+              <TableHeadCell>{vc.colPages}</TableHeadCell>
+            </TableRow>
+          </TableHead>
+          <TableBody>
+            {rows.map((row) => (
+              <TableRow key={`${title}-${row.value}`}>
+                <TableCell className="text-sm break-all">{row.value || '—'}</TableCell>
+                <TableCell>
+                  <div className="flex flex-wrap gap-1">
+                    {(row.sources || []).map((s) => (
+                      <Badge key={s} variant="info" value={s} className="normal-case text-[10px]" />
+                    ))}
+                  </div>
+                </TableCell>
+                <TableCell>
+                  <ul className="space-y-1">
+                    {(row.urls || []).slice(0, 3).map((url) => (
+                      <li key={url}>
+                        <a
+                          href={url}
+                          target="_blank"
+                          rel="noopener noreferrer"
+                          className="text-link text-xs hover:underline break-all inline-flex items-center gap-1"
+                        >
+                          {url}
+                          <ExternalLink className="h-3 w-3 shrink-0" aria-hidden />
+                        </a>
+                      </li>
+                    ))}
+                    {(row.urls?.length ?? 0) > 3 ? (
+                      <li className="text-xs text-muted-foreground">
+                        {format(vc.morePages, { count: (row.urls?.length ?? 0) - 3 })}
+                      </li>
+                    ) : null}
+                  </ul>
+                </TableCell>
+              </TableRow>
+            ))}
+          </TableBody>
+        </Table>
+      </div>
+    </Card>
+  );
+}
+
+export default function Contacts({ searchQuery = '' }: ViewProps) {
+  const { data } = useReport();
+  const vc = strings.views.contacts;
+  const intel = data?.contact_intelligence;
+  const q = (searchQuery || '').toLowerCase().trim();
+
+  const sections = useMemo(
+    (): Record<ContactSection, ContactIntelligenceEntry[]> => ({
+      emails: filterEntries(intel?.emails, q),
+      phones: filterEntries(intel?.phones, q),
+      addresses: filterEntries(intel?.addresses, q),
+      organization_names: filterEntries(intel?.organization_names, q),
+    }),
+    [intel, q],
+  );
+
+  const totalSignals =
+    (intel?.emails?.length ?? 0) +
+    (intel?.phones?.length ?? 0) +
+    (intel?.addresses?.length ?? 0) +
+    (intel?.organization_names?.length ?? 0);
+
+  if (!intel || totalSignals === 0) {
+    return (
+      <PageLayout>
+        <PageHeader title={vc.title} subtitle={vc.subtitle} icon={<Contact2 className="h-7 w-7 text-link shrink-0" />} />
+        <Card>
+          <p className="text-sm text-muted-foreground">{vc.emptyHint}</p>
+        </Card>
+      </PageLayout>
+    );
+  }
+
+  const filteredTotal =
+    sections.emails.length +
+    sections.phones.length +
+    sections.addresses.length +
+    sections.organization_names.length;
+
+  return (
+    <PageLayout>
+      <PageHeader title={vc.title} subtitle={vc.subtitle} icon={<Contact2 className="h-7 w-7 text-link shrink-0" />} />
+      {(intel.consistency_notes?.length ?? 0) > 0 ? (
+        <Card className="mb-6 border-amber-500/30 bg-amber-500/5">
+          <h3 className="text-sm font-semibold text-foreground mb-2">{vc.notesTitle}</h3>
+          <ul className="list-disc list-inside text-sm text-muted-foreground space-y-1">
+            {(intel.consistency_notes || []).map((note) => (
+              <li key={note}>{note}</li>
+            ))}
+          </ul>
+        </Card>
+      ) : null}
+      {intel.primary_contact_page ? (
+        <Card className="mb-6">
+          <h3 className="text-sm font-semibold text-foreground mb-2">{vc.primaryPageTitle}</h3>
+          <a
+            href={intel.primary_contact_page}
+            target="_blank"
+            rel="noopener noreferrer"
+            className="text-link text-sm hover:underline break-all inline-flex items-center gap-1"
+          >
+            {intel.primary_contact_page}
+            <ExternalLink className="h-3.5 w-3.5 shrink-0" aria-hidden />
+          </a>
+        </Card>
+      ) : null}
+      {q && filteredTotal === 0 ? (
+        <Card className="mb-6">
+          <p className="text-sm text-muted-foreground">{vc.noSearchResults}</p>
+        </Card>
+      ) : null}
+      <ContactSectionTable title={vc.emailsTitle} rows={sections.emails} emptyLabel={vc.noEmails} />
+      <ContactSectionTable title={vc.phonesTitle} rows={sections.phones} emptyLabel={vc.noPhones} />
+      <ContactSectionTable title={vc.addressesTitle} rows={sections.addresses} emptyLabel={vc.noAddresses} />
+      <ContactSectionTable
+        title={vc.organizationsTitle}
+        rows={sections.organization_names}
+        emptyLabel={vc.noOrganizations}
+      />
+      <p className="text-xs text-muted-foreground">{vc.provenanceHint}</p>
+    </PageLayout>
+  );
+}
diff --git a/web/src/views/Content.tsx b/web/src/views/Content.tsx
index dcba682e..081e0e2b 100644
--- a/web/src/views/Content.tsx
+++ b/web/src/views/Content.tsx
@@ -11,6 +11,8 @@ import { palette } from '../utils/chartPalette';
 import { registerChartJsBase, barOptionsHorizontal } from '../utils/chartJsDefaults';
 import { formatPageHrefLines } from '../utils/linkUtils';
 import type { ContentUrlEntry, ContentUrlsMap, ViewProps } from '@/types';
+import AiSuggestionButton from '@/components/ai/AiSuggestionButton';
+import { buildDuplicateClusterContext } from '@/lib/fixSuggestionContext';
 
 registerChartJsBase();
 
@@ -150,11 +152,12 @@ export default function Content({ searchQuery = '' }: ViewProps) {
                       <TableHeadCell>{vc.colCluster}</TableHeadCell>
                       <TableHeadCell>{vc.colRepresentative}</TableHeadCell>
                       <TableHeadCell className="text-right">{vc.colUrls}</TableHeadCell>
+                      <TableHeadCell className="w-36" />
                     </tr>
                   </TableHead>
                   <TableBody striped>
                     {(data.content_duplicates || []).slice(0, 40).map((g) => (
-                      <TableRow key={g.id}>
+                      <TableRow key={g.id} className="align-top">
                         <TableCell className="font-mono text-xs text-violet-800 dark:text-violet-300">{g.id}</TableCell>
                         <TableCell className="max-w-md">
                           <a
@@ -169,6 +172,9 @@ export default function Content({ searchQuery = '' }: ViewProps) {
                         <TableCell className="text-right text-muted-foreground text-xs tabular-nums">
                           {g.member_count ?? (g.member_urls || []).length}
                         </TableCell>
+                        <TableCell>
+                          <AiSuggestionButton request={buildDuplicateClusterContext(g)} />
+                        </TableCell>
                       </TableRow>
                     ))}
                   </TableBody>
diff --git a/web/src/views/ContentAnalytics.tsx b/web/src/views/ContentAnalytics.tsx
index 54eda50d..b0bd80b1 100644
--- a/web/src/views/ContentAnalytics.tsx
+++ b/web/src/views/ContentAnalytics.tsx
@@ -10,6 +10,7 @@ import type {
   ResponseTimeStats,
   SeoHealthStats,
   SocialCoverageStats,
+  RichResultsValidationRow,
   ThinPageEntry,
   TopicCluster,
   ViewProps,
@@ -44,6 +45,7 @@ import { statusDistributionFromSummary } from '../lib/statusDistribution';
 import { filterZeroSlices, doughnutOptionsWithPercentTooltip, formatCompositionAria } from '../lib/chartDoughnutUtils';
 import type { ViewTabItem } from '../components';
 import { palette, PALETTE_CATEGORICAL } from '../utils/chartPalette';
+import RichResultsValidationPanel from '../components/content/RichResultsValidationPanel';
 import {
   getGridColor,
   getChartTitleColor,
@@ -316,6 +318,11 @@ export default function ContentAnalytics({ searchQuery = '' }: ViewProps) {
     return arr.filter((u) => String(u).toLowerCase().includes(q));
   }, [data?.social_coverage?.missing_twitter, q]);
 
+  const richResultsRows = useMemo((): RichResultsValidationRow[] => {
+    const raw = data?.rich_results_validation;
+    return Array.isArray(raw) ? raw : [];
+  }, [data?.rich_results_validation]);
+
   const languageMlChart = useMemo(() => {
     const c = data?.language_summary?.counts || {};
     const entries = Object.entries(c)
@@ -561,6 +568,10 @@ export default function ContentAnalytics({ searchQuery = '' }: ViewProps) {
         </Card>
       </div>
 
+      {richResultsRows.length > 0 ? (
+        <RichResultsValidationPanel rows={richResultsRows} meta={data?.rich_results_meta} />
+      ) : null}
+
       {((hreflang?.pages_200 ?? 0) > 0 || outboundDomains.length > 0) && (
         <div className="grid grid-cols-1 lg:grid-cols-2 gap-6">
           {(hreflang?.pages_200 ?? 0) > 0 && (
diff --git a/web/src/views/ExportReport.tsx b/web/src/views/ExportReport.tsx
index 678cd918..acb95a20 100644
--- a/web/src/views/ExportReport.tsx
+++ b/web/src/views/ExportReport.tsx
@@ -4,7 +4,7 @@ import { useCallback, useRef, useState } from 'react';
 import { Download, FileText, Printer } from 'lucide-react';
 import Button from '@/components/Button';
 import { useReport } from '@/context/useReport';
-import { buildAuditExportUrl } from '@/lib/exportAudit';
+import { buildAuditExportUrl, buildWorkbookExportUrl, buildSitemapExportUrl } from '@/lib/exportAudit';
 import { strings } from '@/lib/strings';
 import type { ViewProps } from '@/types/report';
 
@@ -20,6 +20,8 @@ export default function ExportReport(_props: ViewProps) {
   const pdfUrl = buildAuditExportUrl('pdf', reportId);
   const csvUrl = buildAuditExportUrl('csv', reportId);
   const jsonUrl = buildAuditExportUrl('json', reportId);
+  const workbookUrl = buildWorkbookExportUrl(reportId);
+  const sitemapUrl = buildSitemapExportUrl(reportId);
 
   const siteLabel = data?.site_name || strings.app.defaultSiteName;
   const generated = data?.report_generated_at;
@@ -60,6 +62,22 @@ export default function ExportReport(_props: ViewProps) {
               <FileText className="h-4 w-4" />
               {ve.downloadCsv}
             </a>
+            <a
+              href={workbookUrl}
+              download
+              className="inline-flex items-center gap-2 px-3 py-2 rounded-lg text-sm font-medium border border-default text-foreground hover:bg-brand-700/80 transition-colors"
+            >
+              <Download className="h-4 w-4" />
+              {ve.downloadWorkbook}
+            </a>
+            <a
+              href={sitemapUrl}
+              download
+              className="inline-flex items-center gap-2 px-3 py-2 rounded-lg text-sm font-medium border border-default text-foreground hover:bg-brand-700/80 transition-colors"
+            >
+              <Download className="h-4 w-4" />
+              {ve.downloadSitemap}
+            </a>
             <a
               href={jsonUrl}
               className="inline-flex items-center gap-2 px-3 py-2 rounded-lg text-sm font-medium border border-default text-foreground hover:bg-brand-700/80 transition-colors"
diff --git a/web/src/views/Issues.tsx b/web/src/views/Issues.tsx
index aeb66e53..d2510194 100644
--- a/web/src/views/Issues.tsx
+++ b/web/src/views/Issues.tsx
@@ -221,8 +221,11 @@ export default function Issues({ searchQuery = '' }: ViewProps) {
   }
 
   filtered.sort((a, b) => {
-    const aClicks = clicksByUrl.get(String(a.issue.url || '').replace(/\/$/, '')) || 0;
-    const bClicks = clicksByUrl.get(String(b.issue.url || '').replace(/\/$/, '')) || 0;
+    const aImpact = Number(a.issue.impact_score) || 0;
+    const bImpact = Number(b.issue.impact_score) || 0;
+    if (bImpact !== aImpact) return bImpact - aImpact;
+    const aClicks = Number(a.issue.gsc_clicks) || clicksByUrl.get(String(a.issue.url || '').replace(/\/$/, '')) || 0;
+    const bClicks = Number(b.issue.gsc_clicks) || clicksByUrl.get(String(b.issue.url || '').replace(/\/$/, '')) || 0;
     if (bClicks !== aClicks) return bClicks - aClicks;
     const ao = PRIORITY_CONFIG[normalizePriority(a.issue.priority)].order;
     const bo = PRIORITY_CONFIG[normalizePriority(b.issue.priority)].order;
diff --git a/web/src/views/JavaScriptErrors.tsx b/web/src/views/JavaScriptErrors.tsx
index bae59ded..0f3c6be9 100644
--- a/web/src/views/JavaScriptErrors.tsx
+++ b/web/src/views/JavaScriptErrors.tsx
@@ -19,6 +19,8 @@ import {
   linksInspectHref,
   type FlatBrowserErrorRow,
 } from '@/lib/browserErrors';
+import AiSuggestionButton from '@/components/ai/AiSuggestionButton';
+import { buildBrowserErrorContext, buildBrowserErrorSummaryContext } from '@/lib/fixSuggestionContext';
 
 type TypeFilter = 'All' | 'console' | 'exception';
 const JS_ERRORS_TABS = ['summary', 'errors'] as const;
@@ -160,7 +162,14 @@ export default function JavaScriptErrors({ searchQuery = '' }: ViewProps) {
                   <TableBody>
                     {topMessages.map((row) => (
                       <TableRow key={row.text} className="align-top">
-                        <TableCell className="font-mono text-xs break-all">{row.text}</TableCell>
+                        <TableCell className="font-mono text-xs break-all">
+                          <div className="space-y-2">
+                            <span>{row.text}</span>
+                            <AiSuggestionButton
+                              request={buildBrowserErrorSummaryContext(row.text, row.sample_urls, row.count)}
+                            />
+                          </div>
+                        </TableCell>
                         <TableCell className="text-muted-foreground">{row.count.toLocaleString()}</TableCell>
                         <TableCell>
                           <ul className="space-y-1">
@@ -259,7 +268,12 @@ export default function JavaScriptErrors({ searchQuery = '' }: ViewProps) {
                             <TableCell className="text-xs text-muted-foreground capitalize">
                               {row.type === 'console' ? vj.typeConsole : vj.typeException}
                             </TableCell>
-                            <TableCell className="font-mono text-xs break-all">{row.message}</TableCell>
+                            <TableCell className="font-mono text-xs break-all">
+                              <div className="space-y-2">
+                                <span>{row.message}</span>
+                                <AiSuggestionButton request={buildBrowserErrorContext(row)} />
+                              </div>
+                            </TableCell>
                             <TableCell className="font-mono text-xs text-muted-foreground break-all">
                               {formatBrowserErrorSource(row.source_url, row.line)}
                             </TableCell>
diff --git a/web/src/views/KeywordsExplorer.tsx b/web/src/views/KeywordsExplorer.tsx
index 95745cab..40165af3 100644
--- a/web/src/views/KeywordsExplorer.tsx
+++ b/web/src/views/KeywordsExplorer.tsx
@@ -7,6 +7,7 @@ import { Key, Settings2, Play } from 'lucide-react';
 import { useRouter } from 'next/navigation';
 import { useUrlTab } from '@/hooks/useUrlTab';
 import { useReport } from '../context/useReport';
+import { useOptionalPipeline } from '../context/PipelineContext';
 import { useKeywordBrandQuery } from '@/hooks/useKeywordBrandQuery';
 import { filterKeywordRowsForDomain } from '@/lib/filterKeywordsForDomain';
 import { apiUrl } from '../lib/publicBase';
@@ -27,6 +28,10 @@ import {
   ByPagePanel,
   BulkSeedPanel,
 } from '../components/keywordsExplorer/KeywordPanels';
+import TopicMapPanel from '../components/keywordsExplorer/TopicMapPanel';
+import ContentTemplatesPanel from '../components/keywordsExplorer/ContentTemplatesPanel';
+import CompetitorKeywordImport from '../components/keywordsExplorer/CompetitorKeywordImport';
+import CompetitorKeywordGapPanel from '../components/keywordsExplorer/CompetitorKeywordGapPanel';
 import KeywordOverviewPanel from '../components/keywordsExplorer/KeywordOverviewPanel';
 import KeywordTabBanner from '../components/keywordsExplorer/KeywordTabBanner';
 import KeywordFiltersBar from '../components/keywordsExplorer/KeywordFiltersBar';
@@ -42,7 +47,7 @@ import {
   isTableTab,
 } from '../components/keywordsExplorer/keywordTabMeta';
 
-const KEYWORD_TABS = ['overview', ...KEYWORD_TABLE_TAB_IDS, 'cannib', 'alignment', 'bypage'] as const;
+const KEYWORD_TABS = ['overview', ...KEYWORD_TABLE_TAB_IDS, 'cannib', 'alignment', 'bypage', 'topics', 'templates', 'competitor'] as const;
 
 const EMPTY_ROWS: KeywordRow[] = [];
 const EMPTY_HISTORY: KeywordHistoryMap = {};
@@ -50,6 +55,8 @@ const EMPTY_HISTORY: KeywordHistoryMap = {};
 export default function KeywordsExplorer({ onOpenIntegrations }: ViewProps) {
   const router = useRouter();
   const { data, startUrlByRunId, selectedReportId } = useReport();
+  const pipeline = useOptionalPipeline();
+  const propertyId = Number(pipeline?.configState.active_property_id || 0);
   const ke = strings.views.keywordsExplorer;
   const brandQuery = useKeywordBrandQuery();
   const kwData: KeywordReportData | undefined = data?.keywords;
@@ -86,7 +93,12 @@ export default function KeywordsExplorer({ onOpenIntegrations }: ViewProps) {
   const hasGscConnected = !!data?.google?.gsc;
   const showParentTopic = rows.some((r) => r.parent_topic);
   const showTrend = rows.some((r) => r.trend);
-  const tabLabels = ke.tabs as Record<KeywordTabId, string>;
+  const tabLabels = {
+    ...(ke.tabs as Record<string, string>),
+    topics: 'Topic map',
+    templates: 'Content templates',
+    competitor: 'Competitor keywords',
+  } as Record<KeywordTabId, string>;
 
   const quickWinCount = useMemo(
     () =>
@@ -420,6 +432,23 @@ export default function KeywordsExplorer({ onOpenIntegrations }: ViewProps) {
             <QueryPageMisalignmentPanel items={alignmentItems} />
           ) : activeTab === 'bypage' ? (
             <ByPagePanel rows={rows} ke={ke} brandQuery={brandQuery} />
+          ) : activeTab === 'topics' ? (
+            <TopicMapPanel
+              clusters={(data?.semantic_keyword_clusters as Array<{ topic?: string; keywords?: string[] }>) || []}
+              emptyLabel="Run a report with LLM keyword clusters enabled to see topic groups."
+            />
+          ) : activeTab === 'templates' ? (
+            <ContentTemplatesPanel
+              defaultKeyword={rows[0]?.keyword || ''}
+              clusterRows={rows.slice(0, 20)}
+            />
+          ) : activeTab === 'competitor' ? (
+            <>
+              <CompetitorKeywordImport propertyId={propertyId} />
+              <CompetitorKeywordGapPanel
+                rows={Array.isArray(data?.competitor_keyword_gap) ? data.competitor_keyword_gap : []}
+              />
+            </>
           ) : tableEmptyContent ? (
             tableEmptyContent
           ) : (
diff --git a/web/src/views/Links.tsx b/web/src/views/Links.tsx
index 5f5c8d4a..ec403441 100644
--- a/web/src/views/Links.tsx
+++ b/web/src/views/Links.tsx
@@ -1,10 +1,12 @@
 'use client';
 
 import { useState, useMemo, useEffect, useRef, useCallback, type MouseEvent } from 'react';
-import { Link as LinkIcon, ArrowLeft, AlertTriangle, Download } from 'lucide-react';
+import { Link as LinkIcon, ArrowLeft, AlertTriangle, Download, List, TextQuote } from 'lucide-react';
 import { useReport } from '../context/useReport';
 import { strings } from '../lib/strings';
-import { PageLayout, PageHeader, Card, Button, AlertBanner } from '../components';
+import { PageLayout, PageHeader, Card, Button, AlertBanner, ViewTabs } from '../components';
+import type { ViewTabItem } from '../components';
+import { useUrlTab } from '@/hooks/useUrlTab';
 import { useSearchParams, useRouter, usePathname } from 'next/navigation';
 import type {
   InspectorBrokenItem,
@@ -22,6 +24,7 @@ import type {
 import { CopyBtn, InspectorTabs } from '../components/links';
 import {
   type LinkSortKey,
+  LinksExplorerAnchorsTab,
   LinksExplorerTableTab,
 } from '../components/links/explorer';
 import {
@@ -32,6 +35,12 @@ import type { LinksFilterValues } from '../components/links/LinksFilterBar';
 import { linkHasBrowserErrors } from '@/lib/browserErrors';
 import { browserInspectorIssueRows } from '@/components/browser/BrowserDiagnosticsPanel';
 import { exportLinksCsv } from '@/utils/linkExport';
+import { useOptionalPipeline } from '../context/PipelineContext';
+import AiSuggestionButton from '@/components/ai/AiSuggestionButton';
+import { buildTechnicalLinkIssueContext } from '@/lib/fixSuggestionContext';
+
+const EXPLORER_TABS = ['urls', 'anchors'] as const;
+type ExplorerTabId = (typeof EXPLORER_TABS)[number];
 
 const INSPECTOR_TABS = [
   'overview',
@@ -64,6 +73,8 @@ export default function Links({ searchQuery = '' }: ViewProps) {
   const vl = strings.views.links;
   const sj = strings.common;
   const { data } = useReport();
+  const pipeline = useOptionalPipeline();
+  const propertyId = Number(pipeline?.configState.active_property_id || 0);
   const searchParams = useSearchParams();
   const router = useRouter();
   const pathname = usePathname();
@@ -87,6 +98,55 @@ export default function Links({ searchQuery = '' }: ViewProps) {
 
   const links = useMemo(() => data?.links || [], [data]);
 
+  const hasLinkAttributes = Boolean(
+    data?.link_rel_summary || (data?.inlink_anchor_matrix?.length ?? 0) > 0,
+  );
+
+  const explorerValidTabs = useMemo((): readonly ExplorerTabId[] => {
+    if (hasLinkAttributes) return EXPLORER_TABS;
+    return ['urls'];
+  }, [hasLinkAttributes]);
+
+  const [explorerTab, setExplorerTab] = useUrlTab(explorerValidTabs, 'urls');
+
+  const linkAttributeLabels = useMemo(
+    () => ({
+      title: vl.linkAttributesTitle ?? 'Link attributes',
+      total: vl.linkAttrTotal ?? 'Total links',
+      internal: vl.linkAttrInternal ?? 'Internal',
+      nofollow: vl.linkAttrNofollow ?? 'Nofollow internal',
+      sponsored: vl.linkAttrSponsored ?? 'Sponsored internal',
+      external: vl.linkAttrExternal ?? 'External',
+      anchorMatrix: vl.inlinkAnchorMatrix ?? 'Inlink anchor text',
+      target: vl.inlinkTarget ?? 'Target URL',
+      anchor: vl.inlinkAnchor ?? 'Anchor text',
+      inlinks: vl.inlinkCount ?? 'Inlinks',
+      follow: vl.linkAttrFollow ?? 'Follow internal',
+      ugc: vl.linkAttrUgc ?? 'UGC internal',
+    }),
+    [vl],
+  );
+
+  const explorerTabItems = useMemo((): ViewTabItem[] => {
+    const items: ViewTabItem[] = [
+      {
+        id: 'urls',
+        label: vl.tabs.urls,
+        icon: <List className="h-3.5 w-3.5 shrink-0" aria-hidden />,
+        badge: links.length > 0 ? links.length : null,
+      },
+    ];
+    if (hasLinkAttributes) {
+      items.push({
+        id: 'anchors',
+        label: vl.tabs.anchors,
+        icon: <TextQuote className="h-3.5 w-3.5 shrink-0" aria-hidden />,
+        badge: data?.inlink_anchor_matrix?.length ?? null,
+      });
+    }
+    return items;
+  }, [vl.tabs, links.length, hasLinkAttributes, data?.inlink_anchor_matrix?.length]);
+
   const inspectParam = searchParams.get('inspect');
   const tabParam = searchParams.get('tab');
 
@@ -118,13 +178,6 @@ export default function Links({ searchQuery = '' }: ViewProps) {
     [router, pathname, searchParams],
   );
 
-  useEffect(() => {
-    if (inInspector || tabParam !== 'charts') return;
-    replaceParams((params) => {
-      params.delete('tab');
-    });
-  }, [inInspector, tabParam, replaceParams]);
-
   useEffect(() => {
     if (!inspectParam) {
       setInspectNotFound(false);
@@ -307,6 +360,27 @@ export default function Links({ searchQuery = '' }: ViewProps) {
     } as InspectorDetails;
   }, [inspectorUrl, data, links]);
 
+  const siteTechnicalIssues = useMemo(() => {
+    const q = (searchQuery || '').toLowerCase().trim();
+    const issues = data?.issues || {};
+    const rows: Array<{ message: string; url: string; kind: string }> = [];
+    (issues.broken || []).forEach((item) => {
+      const url = String(item.url || '');
+      const message = `Broken link (${item.status ?? 'error'})`;
+      if (!q || `${url} ${message}`.toLowerCase().includes(q)) {
+        rows.push({ message, url, kind: 'broken_link' });
+      }
+    });
+    (issues.redirects || []).forEach((item) => {
+      const url = String(item.url || '');
+      const message = `Redirect ${item.status ?? ''} → ${item.final_url || ''}`.trim();
+      if (!q || `${url} ${message}`.toLowerCase().includes(q)) {
+        rows.push({ message, url, kind: 'redirect' });
+      }
+    });
+    return rows.slice(0, 40);
+  }, [data?.issues, searchQuery]);
+
   const handleRowMouseEnter = useCallback((e: MouseEvent<HTMLTableRowElement>, link: ReportLink) => {
     const rect = e.currentTarget.getBoundingClientRect();
     const containerRect = tableRef.current?.getBoundingClientRect?.() || { top: 0, left: 0, width: 800 };
@@ -358,6 +432,15 @@ export default function Links({ searchQuery = '' }: ViewProps) {
     setPage(1);
   };
 
+  const loadSavedFilter = (values: LinksFilterValues) => {
+    setInlinksFilter(values.inlinksFilter);
+    setStatusFilter(values.statusFilter);
+    setRtFilter(values.rtFilter);
+    setWcFilter(values.wcFilter);
+    setJsErrorFilter(values.jsErrorFilter);
+    setPage(1);
+  };
+
   const linkForInspector = inspectorUrl ? (links.find((l) => l.url === inspectorUrl) || null) : null;
 
   return (
@@ -387,7 +470,7 @@ export default function Links({ searchQuery = '' }: ViewProps) {
             }
             className="mb-0"
             actions={
-              filtered.length > 0 ? (
+              explorerTab === 'urls' && filtered.length > 0 ? (
                 <Button
                   type="button"
                   variant="secondary"
@@ -401,10 +484,37 @@ export default function Links({ searchQuery = '' }: ViewProps) {
             }
           />
 
+          <ViewTabs
+            tabs={explorerTabItems}
+            activeTab={explorerTab}
+            onChange={(id) => setExplorerTab(id as ExplorerTabId)}
+            ariaLabel={vl.title}
+            idPrefix="links-explorer"
+          />
+
+          {siteTechnicalIssues.length > 0 ? (
+            <Card className="p-4 space-y-3">
+              <h2 className="text-sm font-bold text-foreground">{vl.siteTechnicalIssuesTitle}</h2>
+              <p className="text-xs text-muted-foreground">{vl.siteTechnicalIssuesHint}</p>
+              <ul className="space-y-3 max-h-64 overflow-y-auto">
+                {siteTechnicalIssues.map((row, i) => (
+                  <li key={`${row.url}-${row.kind}-${i}`} className="border border-default rounded-lg px-3 py-2 space-y-2">
+                    <div className="text-xs font-mono text-link break-all">{row.url}</div>
+                    <div className="text-xs text-muted-foreground">{row.message}</div>
+                    <AiSuggestionButton request={buildTechnicalLinkIssueContext(row.message, row.url, row.kind)} />
+                  </li>
+                ))}
+              </ul>
+            </Card>
+          ) : null}
+
+          {explorerTab === 'urls' ? (
           <LinksExplorerTableTab
             filterValues={filterValues}
             onFilterChange={handleFilterChange}
             onClearAllFilters={clearAllFilters}
+            propertyId={propertyId}
+            onLoadSavedFilter={loadSavedFilter}
             searchQuery={searchQuery}
             filtered={filtered}
             pageLinks={pageLinks}
@@ -424,6 +534,13 @@ export default function Links({ searchQuery = '' }: ViewProps) {
             onRowMouseEnter={handleRowMouseEnter}
             onRowMouseLeave={() => setHoveredRow(null)}
           />
+          ) : (
+            <LinksExplorerAnchorsTab
+              summary={data?.link_rel_summary}
+              anchors={data?.inlink_anchor_matrix}
+              labels={linkAttributeLabels}
+            />
+          )}
         </>
       ) : (
         <>
diff --git a/web/src/views/Redirects.tsx b/web/src/views/Redirects.tsx
index 14b65e53..ffb1bd45 100644
--- a/web/src/views/Redirects.tsx
+++ b/web/src/views/Redirects.tsx
@@ -7,6 +7,8 @@ import { PageLayout, PageHeader, Card, Table, TableHead, TableHeadCell, TableBod
 import { palette } from '../utils/chartPalette';
 import { registerChartJsBase, barOptionsHorizontal } from '../utils/chartJsDefaults';
 import type { ReportRedirect, ViewProps } from '@/types';
+import AiSuggestionButton from '@/components/ai/AiSuggestionButton';
+import { buildRedirectContext } from '@/lib/fixSuggestionContext';
 
 registerChartJsBase();
 
@@ -84,11 +86,12 @@ export default function Redirects({ searchQuery = '' }: ViewProps) {
                 <TableHeadCell>{vr.colFrom}</TableHeadCell>
                 <TableHeadCell>{vr.colStatus}</TableHeadCell>
                 <TableHeadCell>{vr.colTo}</TableHeadCell>
+                <TableHeadCell className="w-36" />
               </tr>
             </TableHead>
             <TableBody>
               {redirects.map((r, i) => (
-                <TableRow key={i}>
+                <TableRow key={i} className="align-top">
                   <TableCell className="font-mono text-link text-xs break-all py-3">
                     <a href={r.url || r.from} target="_blank" rel="noreferrer" className="hover:underline">
                       {r.url || r.from}
@@ -102,6 +105,9 @@ export default function Redirects({ searchQuery = '' }: ViewProps) {
                       {r.final_url || r.to}
                     </a>
                   </TableCell>
+                  <TableCell className="py-3">
+                    <AiSuggestionButton request={buildRedirectContext(r)} />
+                  </TableCell>
                 </TableRow>
               ))}
             </TableBody>
diff --git a/web/src/views/Security.tsx b/web/src/views/Security.tsx
index 65bedd28..084ffe05 100644
--- a/web/src/views/Security.tsx
+++ b/web/src/views/Security.tsx
@@ -13,6 +13,8 @@ import { doughnutOptionsWithPercentTooltip, formatCompositionAria } from '../lib
 import { ChartAccessibleFallback } from '../components/charts';
 import type { SecurityFinding, ViewProps } from '@/types';
 import { securityFindingLabel } from '@/lib/securityFindingLabels';
+import AiSuggestionButton from '@/components/ai/AiSuggestionButton';
+import { buildSecurityFindingContext } from '@/lib/fixSuggestionContext';
 
 registerChartJsBase();
 
@@ -355,6 +357,7 @@ export default function Security({ searchQuery = '' }: ViewProps) {
                         {f.recommendation}
                       </div>
                     )}
+                    <AiSuggestionButton request={buildSecurityFindingContext(f)} />
                   </div>
                 );
               })}
diff --git a/web/src/views/SiteStructure.tsx b/web/src/views/SiteStructure.tsx
index 907a45fd..ba92cc54 100644
--- a/web/src/views/SiteStructure.tsx
+++ b/web/src/views/SiteStructure.tsx
@@ -1,3 +1,4 @@
+import dynamic from 'next/dynamic';
 import { useMemo, useState, useCallback, useEffect } from 'react';
 import { useUrlTab } from '@/hooks/useUrlTab';
 import {
@@ -10,6 +11,7 @@ import {
   ChevronsDownUp,
   ChevronsUpDown,
   BarChart3,
+  Share2,
 } from 'lucide-react';
 import { useReport } from '../context/useReport';
 import { strings, format } from '../lib/strings';
@@ -22,16 +24,24 @@ import {
   defaultExpandedPathKeys,
   filterLinksBySearch,
   finalizeRollup,
+  linkMatchesPathKey,
 } from '../lib/siteStructureTree';
 import { PageLayout, PageHeader, Card, Button, StatCard, AlertBanner, ViewTabs, ViewTabPanel } from '../components';
 import UrlInspectorButton from '@/components/UrlInspectorButton';
 import type { ViewTabItem } from '../components';
 import PathTreeTable from '../components/siteStructure/PathTreeTable';
+import CrawlMapPanel from '../components/siteStructure/CrawlMapPanel';
+
+const SiteStructureLinkGraph = dynamic(
+  () => import('../components/siteStructure/SiteStructureLinkGraph'),
+  { ssr: false, loading: () => <p className="text-sm text-muted-foreground py-8 text-center">Loading link graph…</p> },
+);
+
 import type { CrawlSegmentEntry, CrawlSegmentsData, PathTreeNode, PathTreeTableRow, ViewProps } from '@/types';
 
 const TREE_PAGE_SIZE = 20;
 
-const SITE_STRUCTURE_TABS = ['overview', 'tree'] as const;
+const SITE_STRUCTURE_TABS = ['overview', 'tree', 'map', 'graph'] as const;
 type SiteStructureTabId = (typeof SITE_STRUCTURE_TABS)[number];
 
 interface SiteStructureTreePanelProps {
@@ -240,6 +250,7 @@ export default function SiteStructure({ searchQuery = '' }: ViewProps) {
   const s = strings.views.siteStructure;
   const { data, compareData, startUrlByRunId, selectedReportId, compareReportId } = useReport();
   const [showCompareCharts, setShowCompareCharts] = useState(true);
+  const [pathPrefixFilter, setPathPrefixFilter] = useState<string | null>(null);
   const [activeTab, setActiveTab] = useUrlTab(SITE_STRUCTURE_TABS, 'overview');
 
   const expectedHost = useMemo(
@@ -247,15 +258,21 @@ export default function SiteStructure({ searchQuery = '' }: ViewProps) {
     [data, startUrlByRunId]
   );
 
-  const filteredLinks = useMemo(
-    () => filterLinksBySearch(data?.links || [], searchQuery),
-    [data?.links, searchQuery]
-  );
+  const filteredLinks = useMemo(() => {
+    let links = filterLinksBySearch(data?.links || [], searchQuery);
+    if (pathPrefixFilter) {
+      links = links.filter((l) => linkMatchesPathKey(String(l.url || ''), pathPrefixFilter, expectedHost));
+    }
+    return links;
+  }, [data?.links, searchQuery, pathPrefixFilter, expectedHost]);
 
-  const baselineLinks = useMemo(
-    () => filterLinksBySearch(compareData?.links || [], searchQuery),
-    [compareData?.links, searchQuery]
-  );
+  const baselineLinks = useMemo(() => {
+    let links = filterLinksBySearch(compareData?.links || [], searchQuery);
+    if (pathPrefixFilter) {
+      links = links.filter((l) => linkMatchesPathKey(String(l.url || ''), pathPrefixFilter, expectedHost));
+    }
+    return links;
+  }, [compareData?.links, searchQuery, pathPrefixFilter, expectedHost]);
 
   const hasCompare = compareData != null && compareReportId != null;
   const searchActive = (searchQuery || '').trim().length > 0;
@@ -311,8 +328,20 @@ export default function SiteStructure({ searchQuery = '' }: ViewProps) {
         icon: <FolderTree className="h-3.5 w-3.5 shrink-0" aria-hidden />,
         badge: filteredLinks.length > 0 ? filteredLinks.length : null,
       },
+      {
+        id: 'map',
+        label: 'Crawl map',
+        icon: <Layers className="h-3.5 w-3.5 shrink-0" aria-hidden />,
+        badge: null,
+      },
+      {
+        id: 'graph',
+        label: 'Link graph',
+        icon: <Share2 className="h-3.5 w-3.5 shrink-0" aria-hidden />,
+        badge: (data?.graph_nodes?.length ?? 0) > 0 ? Math.min(200, data!.graph_nodes!.length) : null,
+      },
     ];
-  }, [s.tabs, merged.size, filteredLinks.length]);
+  }, [s.tabs, merged.size, filteredLinks.length, data?.graph_nodes?.length]);
 
   if (!data) return null;
 
@@ -453,6 +482,51 @@ export default function SiteStructure({ searchQuery = '' }: ViewProps) {
           </Card>
         </ViewTabPanel>
       )}
+
+      {activeTab === 'map' && tree ? (
+        <ViewTabPanel idPrefix="site-structure" tabId="map" className="space-y-3">
+          {pathPrefixFilter ? (
+            <AlertBanner variant="info" className="text-xs py-2 flex flex-wrap items-center gap-2">
+              <span>
+                Filtering tree to <span className="font-mono font-semibold">{pathPrefixFilter}</span>
+              </span>
+              <Button
+                type="button"
+                variant="secondary"
+                className="text-xs py-0.5 px-2"
+                onClick={() => setPathPrefixFilter(null)}
+              >
+                Clear filter
+              </Button>
+              <Button
+                type="button"
+                variant="secondary"
+                className="text-xs py-0.5 px-2"
+                onClick={() => setActiveTab('tree')}
+              >
+                View in tree
+              </Button>
+            </AlertBanner>
+          ) : null}
+          <CrawlMapPanel
+            tree={tree}
+            crawlSegments={crawlSegments}
+            selectedPath={pathPrefixFilter}
+            onSelectPath={(pathKey) => {
+              setPathPrefixFilter(pathKey);
+              setActiveTab('tree');
+            }}
+          />
+        </ViewTabPanel>
+      ) : null}
+
+      {activeTab === 'graph' && data ? (
+        <ViewTabPanel idPrefix="site-structure" tabId="graph">
+          <Card padding="default">
+            <SiteStructureLinkGraph />
+          </Card>
+        </ViewTabPanel>
+      ) : null}
     </PageLayout>
   );
 }
diff --git a/web/src/views/Subdomains.tsx b/web/src/views/Subdomains.tsx
new file mode 100644
index 00000000..e61f9037
--- /dev/null
+++ b/web/src/views/Subdomains.tsx
@@ -0,0 +1,169 @@
+'use client';
+
+import { useMemo } from 'react';
+import Link from 'next/link';
+import { useSearchParams } from 'next/navigation';
+import { Globe2 } from 'lucide-react';
+import { useReport } from '../context/useReport';
+import { strings, format } from '../lib/strings';
+import {
+  PageLayout,
+  PageHeader,
+  Card,
+  StatCard,
+  Badge,
+  Table,
+  TableHead,
+  TableHeadCell,
+  TableBody,
+  TableRow,
+  TableCell,
+} from '../components';
+import type { SubdomainHostEntry, ViewProps } from '@/types';
+
+function yesNo(value: boolean | undefined): string {
+  return value ? strings.common.yes : strings.common.no;
+}
+
+export default function Subdomains({ searchQuery = '' }: ViewProps) {
+  const { data } = useReport();
+  const searchParams = useSearchParams();
+  const vs = strings.views.subdomains;
+  const inv = data?.subdomains;
+  const q = (searchQuery || '').toLowerCase().trim();
+
+  const querySuffix = searchParams.toString() ? `?${searchParams.toString()}` : '';
+
+  const inScopeHosts = useMemo(() => {
+    const all = (inv?.hosts || []).filter((h): h is SubdomainHostEntry & { host: string } => Boolean(h.host));
+    const scoped = all.filter((h) => h.in_scope !== false);
+    if (!q) return scoped;
+    return scoped.filter((h) => {
+      const host = h.host.toLowerCase();
+      const sources = (h.sources || []).join(' ').toLowerCase();
+      return host.includes(q) || sources.includes(q);
+    });
+  }, [inv?.hosts, q]);
+
+  const gscGapHosts = inv?.gsc_hosts_not_crawled || [];
+  const outOfScope = inv?.out_of_scope_discovered || [];
+
+  if (!inv || inv.disabled) {
+    return (
+      <PageLayout>
+        <PageHeader title={vs.title} subtitle={vs.subtitle} icon={<Globe2 className="h-7 w-7 text-link shrink-0" />} />
+        <Card>
+          <p className="text-sm text-muted-foreground">{vs.disabledHint}</p>
+        </Card>
+      </PageLayout>
+    );
+  }
+
+  if (!inv.hosts?.length && !gscGapHosts.length) {
+    return (
+      <PageLayout>
+        <PageHeader title={vs.title} subtitle={vs.subtitle} icon={<Globe2 className="h-7 w-7 text-link shrink-0" />} />
+        <Card>
+          <p className="text-sm text-muted-foreground">{vs.emptyHint}</p>
+        </Card>
+      </PageLayout>
+    );
+  }
+
+  return (
+    <PageLayout>
+      <PageHeader title={vs.title} subtitle={vs.subtitle} icon={<Globe2 className="h-7 w-7 text-link shrink-0" />} />
+      {inv.crtsh_error ? (
+        <Card className="mb-4 border-amber-500/30 bg-amber-500/5">
+          <p className="text-sm text-muted-foreground">{vs.ctWarning}</p>
+        </Card>
+      ) : null}
+      <div className="grid grid-cols-2 sm:grid-cols-4 gap-3 mb-6">
+        <StatCard label={vs.apex} value={inv.apex || '—'} />
+        <StatCard label={vs.inScopeHosts} value={inScopeHosts.length} />
+        <StatCard label={vs.gscNotCrawled} value={gscGapHosts.length} />
+        <StatCard label={vs.outOfScope} value={outOfScope.length} />
+      </div>
+      {gscGapHosts.length > 0 ? (
+        <Card className="mb-6">
+          <h3 className="text-sm font-semibold text-foreground mb-2">{vs.gscGapTitle}</h3>
+          <p className="text-sm text-muted-foreground mb-3">{vs.gscGapHint}</p>
+          <ul className="flex flex-wrap gap-2">
+            {gscGapHosts.slice(0, 20).map((host) => (
+              <li key={host}>
+                <Badge variant="medium" value={host} className="normal-case font-mono text-[11px]" />
+              </li>
+            ))}
+          </ul>
+          {gscGapHosts.length > 20 ? (
+            <p className="text-xs text-muted-foreground mt-2">{format(vs.moreHosts, { count: gscGapHosts.length - 20 })}</p>
+          ) : null}
+          <p className="text-xs text-muted-foreground mt-3">
+            <Link href={`/indexation${querySuffix}`} className="text-link hover:underline">
+              {vs.viewIndexation}
+            </Link>
+          </p>
+        </Card>
+      ) : null}
+      <Card>
+        <h3 className="text-sm font-semibold text-foreground mb-4">{vs.hostsTitle}</h3>
+        <div className="overflow-x-auto">
+          <Table>
+            <TableHead>
+              <TableRow>
+                <TableHeadCell>{vs.colHost}</TableHeadCell>
+                <TableHeadCell>{vs.colSources}</TableHeadCell>
+                <TableHeadCell>{vs.colCrawl}</TableHeadCell>
+                <TableHeadCell>{vs.colGsc}</TableHeadCell>
+                <TableHeadCell>{vs.colCrawlUrls}</TableHeadCell>
+                <TableHeadCell>{vs.colGscUrls}</TableHeadCell>
+              </TableRow>
+            </TableHead>
+            <TableBody>
+              {inScopeHosts.length === 0 ? (
+                <TableRow>
+                  <TableCell className="text-muted-foreground text-sm">{vs.noSearchResults}</TableCell>
+                  <TableCell />
+                  <TableCell />
+                  <TableCell />
+                  <TableCell />
+                  <TableCell />
+                </TableRow>
+              ) : (
+                inScopeHosts.map((row) => (
+                  <TableRow key={row.host}>
+                    <TableCell className="font-mono text-xs">{row.host}</TableCell>
+                    <TableCell>
+                      <div className="flex flex-wrap gap-1">
+                        {(row.sources || []).map((s) => (
+                          <Badge key={s} variant="info" value={s} className="normal-case text-[10px]" />
+                        ))}
+                      </div>
+                    </TableCell>
+                    <TableCell>{yesNo(row.in_crawl)}</TableCell>
+                    <TableCell>{yesNo(row.in_gsc)}</TableCell>
+                    <TableCell>{row.url_count_crawl ?? 0}</TableCell>
+                    <TableCell>{row.url_count_gsc ?? 0}</TableCell>
+                  </TableRow>
+                ))
+              )}
+            </TableBody>
+          </Table>
+        </div>
+      </Card>
+      {outOfScope.length > 0 ? (
+        <Card className="mt-6">
+          <h3 className="text-sm font-semibold text-foreground mb-2">{vs.outOfScopeTitle}</h3>
+          <p className="text-sm text-muted-foreground mb-3">{vs.outOfScopeHint}</p>
+          <ul className="text-sm font-mono text-xs space-y-1 max-h-40 overflow-y-auto">
+            {outOfScope.slice(0, 30).map((host) => (
+              <li key={host} className="text-muted-foreground">
+                {host}
+              </li>
+            ))}
+          </ul>
+        </Card>
+      ) : null}
+    </PageLayout>
+  );
+}