diff --git a/code_review_graph/hints.py b/code_review_graph/hints.py
index 9fd8ce59c..589699626 100644
--- a/code_review_graph/hints.py
+++ b/code_review_graph/hints.py
@@ -12,153 +12,169 @@
 from collections import deque
 from typing import Any
 
-# ---- intent categories and their characteristic tool names ----
+# ---- intent categories and their characteristic public MCP tool names ----
+# Keep these keyed on exported MCP tool names only. Do not substitute
+# internal helper names from implementation modules here.
 
 _INTENT_TOOLS: dict[str, set[str]] = {
     "reviewing": {
-        "detect_changes", "get_review_context", "get_affected_flows", "get_impact_radius",
+        "detect_changes_tool", "get_review_context_tool", "get_affected_flows_tool", "get_impact_radius_tool",
     },
     "debugging": {
-        "query_graph", "get_flow", "semantic_search_nodes",
+        "query_graph_tool", "get_flow_tool", "semantic_search_nodes_tool",
     },
     "refactoring": {
-        "refactor", "find_dead_code", "suggest_refactorings",
+        "refactor_tool", "apply_refactor_tool",
     },
     "exploring": {
-        "list_communities", "get_architecture_overview", "list_flows", "list_graph_stats",
+        "list_communities_tool", "get_architecture_overview_tool", "list_flows_tool", "list_graph_stats_tool",
     },
 }
 
 # ---- workflow adjacency: for each tool, which tools are useful next ----
 
 _WORKFLOW: dict[str, list[dict[str, str]]] = {
-    "list_flows": [
+    "list_flows_tool": [
         {
-            "tool": "get_flow",
+            "tool": "get_flow_tool",
             "suggestion": "Drill into a specific flow for step-by-step details",
         },
         {
-            "tool": "get_affected_flows",
+            "tool": "get_affected_flows_tool",
             "suggestion": "Check which flows are affected by recent changes",
         },
         {
-            "tool": "get_architecture_overview",
+            "tool": "get_architecture_overview_tool",
             "suggestion": "See the high-level architecture",
         },
     ],
-    "get_flow": [
+    "get_flow_tool": [
         {
-            "tool": "query_graph",
+            "tool": "query_graph_tool",
             "suggestion": "Inspect callers/callees of a step in this flow",
         },
         {
-            "tool": "get_affected_flows",
+            "tool": "get_affected_flows_tool",
             "suggestion": "Check if changes affect this flow",
         },
         {
-            "tool": "list_flows",
+            "tool": "list_flows_tool",
             "suggestion": "Browse other execution flows",
         },
     ],
-    "get_affected_flows": [
+    "get_affected_flows_tool": [
         {
-            "tool": "detect_changes",
+            "tool": "detect_changes_tool",
             "suggestion": "Get risk-scored change analysis",
         },
         {
-            "tool": "get_flow",
+            "tool": "get_flow_tool",
             "suggestion": "Inspect a specific affected flow",
         },
         {
-            "tool": "get_review_context",
+            "tool": "get_review_context_tool",
             "suggestion": "Build a full review context for the changes",
         },
     ],
-    "list_communities": [
+    "list_communities_tool": [
         {
-            "tool": "get_community",
+            "tool": "get_community_tool",
             "suggestion": "Inspect a specific community's members",
         },
         {
-            "tool": "get_architecture_overview",
+            "tool": "get_architecture_overview_tool",
             "suggestion": "See cross-community coupling and warnings",
         },
         {
-            "tool": "list_flows",
+            "tool": "list_flows_tool",
             "suggestion": "See execution flows across communities",
         },
     ],
-    "get_community": [
+    "get_community_tool": [
         {
-            "tool": "query_graph",
+            "tool": "query_graph_tool",
             "suggestion": "Explore callers/callees of community members",
         },
         {
-            "tool": "list_communities",
+            "tool": "list_communities_tool",
             "suggestion": "Browse other communities",
         },
         {
-            "tool": "get_architecture_overview",
+            "tool": "get_architecture_overview_tool",
             "suggestion": "See how this community fits the architecture",
         },
     ],
-    "get_architecture_overview": [
+    "get_architecture_overview_tool": [
         {
-            "tool": "list_communities",
+            "tool": "list_communities_tool",
             "suggestion": "Drill into individual communities",
         },
         {
-            "tool": "detect_changes",
+            "tool": "detect_changes_tool",
             "suggestion": "See how recent changes affect the architecture",
         },
         {
-            "tool": "list_flows",
+            "tool": "list_flows_tool",
             "suggestion": "Explore execution flows",
         },
     ],
-    "detect_changes": [
+    "detect_changes_tool": [
         {
-            "tool": "get_review_context",
+            "tool": "get_review_context_tool",
             "suggestion": "Build a full review context with source snippets",
         },
         {
-            "tool": "get_affected_flows",
+            "tool": "get_affected_flows_tool",
             "suggestion": "See which execution flows are affected",
         },
         {
-            "tool": "get_impact_radius",
+            "tool": "get_impact_radius_tool",
             "suggestion": "Expand the blast radius analysis",
         },
         {
-            "tool": "refactor",
+            "tool": "refactor_tool",
             "suggestion": "Look for refactoring opportunities in changed code",
         },
     ],
-    "refactor": [
+    "refactor_tool": [
         {
-            "tool": "query_graph",
+            "tool": "apply_refactor_tool",
+            "suggestion": "Apply a reviewed rename preview",
+        },
+        {
+            "tool": "semantic_search_nodes_tool",
+            "suggestion": "Find related symbols to also rename",
+        },
+        {
+            "tool": "query_graph_tool",
             "suggestion": "Verify call sites before applying a rename",
         },
+    ],
+    "apply_refactor_tool": [
         {
-            "tool": "detect_changes",
-            "suggestion": "Check risk of the refactored code",
+            "tool": "detect_changes_tool",
+            "suggestion": "Check the impact of the applied refactor",
         },
         {
-            "tool": "semantic_search_nodes",
-            "suggestion": "Find related symbols to also rename",
+            "tool": "get_affected_flows_tool",
+            "suggestion": "See which execution flows may have changed",
+        },
+        {
+            "tool": "query_graph_tool",
+            "suggestion": "Inspect remaining call sites or rename fallout",
         },
     ],
-    "semantic_search_nodes": [
+    "semantic_search_nodes_tool": [
         {
-            "tool": "query_graph",
+            "tool": "query_graph_tool",
             "suggestion": "Inspect callers/callees of a search result",
         },
         {
-            "tool": "get_flow",
+            "tool": "get_flow_tool",
             "suggestion": "See the execution flow through a matched node",
         },
         {
-            "tool": "get_impact_radius",
+            "tool": "get_impact_radius_tool",
             "suggestion": "Check the blast radius from matched nodes",
         },
     ],
diff --git a/code_review_graph/prompts.py b/code_review_graph/prompts.py
index 455a4e6b7..1ca9663ca 100644
--- a/code_review_graph/prompts.py
+++ b/code_review_graph/prompts.py
@@ -1,9 +1,9 @@
 """MCP prompt templates for Code Review Graph.
 
 Provides 5 pre-built prompt workflows, all enforcing token-efficient
-detail_level="minimal" first patterns with get_minimal_context entry point.
+detail_level="minimal" first patterns with get_minimal_context_tool entry point.
 
-1. review_changes   - pre-commit review using detect_changes + affected_flows
+1. review_changes   - pre-commit review using detect_changes_tool + get_affected_flows_tool
 2. architecture_map - architecture docs using communities, flows, Mermaid
 3. debug_issue      - guided debugging using search, flow tracing
 4. onboard_developer - new dev orientation using stats, architecture, flows
@@ -17,15 +17,15 @@
 _TOKEN_EFFICIENCY_PREAMBLE = (  # nosec B105 — prompt template, not a password
     """\
 ## Rules for Token-Efficient Graph Usage
-1. ALWAYS call `get_minimal_context` first with a task description.
+1. ALWAYS call `get_minimal_context_tool` first with a task description.
 2. Use `detail_level="minimal"` on all tool calls unless the minimal output \
 is insufficient.
 3. Only escalate to `detail_level="standard"` or `"verbose"` for the specific \
 entities that need deeper inspection.
 4. Never request more than 3 tool calls per turn unless absolutely necessary.
-5. Prefer targeted queries (query_graph with a specific symbol) over broad \
-scans (list_communities with full members).
-6. When reviewing changes: detect_changes(detail_level="minimal") → only \
+5. Prefer targeted queries (query_graph_tool with a specific symbol) over broad \
+scans (list_communities_tool with full members).
+6. When reviewing changes: detect_changes_tool(detail_level="minimal") → only \
 expand on high-risk items.
 """
 )
@@ -51,22 +51,22 @@ def review_changes_prompt(base: str = "HEAD~1") -> list[Message]:
     return _user(
         f"{_TOKEN_EFFICIENCY_PREAMBLE}\n"
         f"## Review Workflow\n"
-        f'1. Call `get_minimal_context(task="review changes against '
+        f'1. Call `get_minimal_context_tool(task="review changes against '
         f'{base}")` to get risk overview.\n'
         f'2. If risk is "low": call '
-        f'`detect_changes(detail_level="minimal")` → report summary '
+        f'`detect_changes_tool(detail_level="minimal")` → report summary '
         f"+ any test gaps.\n"
         f'3. If risk is "medium" or "high":\n'
-        f'   a. Call `detect_changes(detail_level="standard")` for '
+        f'   a. Call `detect_changes_tool(detail_level="standard")` for '
         f"full change list.\n"
         f"   b. For each high-risk function, call "
-        f'`query_graph(pattern="callers_of", target=<func>, '
+        f'`query_graph_tool(pattern="callers_of", target=<func>, '
         f'detail_level="minimal")`.\n'
-        f'   c. Call `get_affected_flows(detail_level="minimal")` '
+        f'   c. Call `get_affected_flows_tool(detail_level="minimal")` '
         f"only if >3 changed functions.\n"
         f"4. Summarize: risk level, what changed, test gaps, "
         f"specific improvements needed.\n\n"
-        f"Do NOT call get_review_context unless you need source code "
+        f"Do NOT call get_review_context_tool unless you need source code "
         f"snippets for a specific function."
     )
 
@@ -76,12 +76,12 @@ def architecture_map_prompt() -> list[Message]:
     return _user(
         f"{_TOKEN_EFFICIENCY_PREAMBLE}\n"
         "## Architecture Mapping Workflow\n"
-        '1. Call `get_minimal_context(task="map architecture")`.\n'
-        '2. Call `get_architecture_overview(detail_level="minimal")` '
+        '1. Call `get_minimal_context_tool(task="map architecture")`.\n'
+        '2. Call `get_architecture_overview_tool(detail_level="minimal")` '
         "for community coupling summary.\n"
-        '3. Call `list_flows(detail_level="minimal")` for critical '
+        '3. Call `list_flows_tool(detail_level="minimal")` for critical '
         "flow names + criticality scores.\n"
-        "4. Only call `get_community(name=<X>, "
+        "4. Only call `get_community_tool(name=<X>, "
         'detail_level="standard")` for the 1-2 communities the user '
         "is most interested in.\n"
         "5. Produce a concise Mermaid diagram showing communities as "
@@ -99,17 +99,17 @@ def debug_issue_prompt(description: str = "") -> list[Message]:
     return _user(
         f"{_TOKEN_EFFICIENCY_PREAMBLE}\n"
         "## Debug Workflow\n"
-        f'1. Call `get_minimal_context(task="debug: '
+        f'1. Call `get_minimal_context_tool(task="debug: '
         f'{desc_part}")`.\n'
-        "2. Call `semantic_search_nodes(query=<keywords from "
+        "2. Call `semantic_search_nodes_tool(query=<keywords from "
         'description>, detail_level="minimal", limit=5)`.\n'
         "3. For the top 1-2 results, call "
-        '`query_graph(pattern="callers_of", target=<name>, '
+        '`query_graph_tool(pattern="callers_of", target=<name>, '
         'detail_level="minimal")`.\n'
         "4. If the issue involves execution flow: call "
-        "`get_flow(name=<relevant flow>)` for the single most "
+        "`get_flow_tool(name=<relevant flow>)` for the single most "
         "relevant flow.\n"
-        "5. Only call `get_review_context` or `get_impact_radius` "
+        "5. Only call `get_review_context_tool` or `get_impact_radius_tool` "
         "if you need to trace the blast radius of a specific change."
     )
 
@@ -119,13 +119,13 @@ def onboard_developer_prompt() -> list[Message]:
     return _user(
         f"{_TOKEN_EFFICIENCY_PREAMBLE}\n"
         "## Onboarding Workflow\n"
-        '1. Call `get_minimal_context(task="onboard developer")`.\n'
-        "2. Call `list_graph_stats()` for technology overview.\n"
-        '3. Call `get_architecture_overview(detail_level="minimal")` '
+        '1. Call `get_minimal_context_tool(task="onboard developer")`.\n'
+        "2. Call `list_graph_stats_tool()` for technology overview.\n"
+        '3. Call `get_architecture_overview_tool(detail_level="minimal")` '
         "for the 30-second mental model.\n"
-        '4. Call `list_communities(detail_level="minimal")` — '
+        '4. Call `list_communities_tool(detail_level="minimal")` — '
         "present as a table of module names + sizes.\n"
-        '5. Call `list_flows(detail_level="minimal")` — highlight '
+        '5. Call `list_flows_tool(detail_level="minimal")` — highlight '
         "the top 3 critical flows.\n"
         "6. Only drill into a specific community or flow if the "
         "developer asks."
@@ -141,19 +141,22 @@ def pre_merge_check_prompt(base: str = "HEAD~1") -> list[Message]:
     return _user(
         f"{_TOKEN_EFFICIENCY_PREAMBLE}\n"
         "## Pre-Merge Check Workflow\n"
-        '1. Call `get_minimal_context(task="pre-merge check")`.\n'
-        '2. Call `detect_changes(detail_level="minimal")` for risk '
+        '1. Call `get_minimal_context_tool(task="pre-merge check")`.\n'
+        '2. Call `detect_changes_tool(detail_level="minimal")` for risk '
         "score and test gaps.\n"
         "3. If risk > 0.4: call "
-        '`get_affected_flows(detail_level="minimal")`.\n'
+        '`get_affected_flows_tool(detail_level="minimal")`.\n'
         "4. If test_gap_count > 0: call "
-        '`query_graph(pattern="tests_for", '
+        '`query_graph_tool(pattern="tests_for", '
         'target=<each untested function>, detail_level="minimal")` '
         "for up to 3 functions.\n"
-        '5. Call `refactor(mode="dead_code", '
+        '5. Call `refactor_tool(mode="dead_code", '
         'detail_level="minimal")` to check for newly dead code.\n'
-        "6. Only call `find_large_functions` or `get_impact_radius` "
+        "6. If you are planning a rename, also call "
+        "`semantic_search_nodes_tool(detail_level=\"minimal\")` to find "
+        "related symbols to also rename.\n"
+        "7. Only call `find_large_functions_tool` or `get_impact_radius_tool` "
         "if risk > 0.7.\n"
-        "7. Output: GO/NO-GO recommendation with 1-sentence "
+        "8. Output: GO/NO-GO recommendation with 1-sentence "
         "justification + list of required follow-ups."
     )
diff --git a/code_review_graph/skills.py b/code_review_graph/skills.py
index 8088e127b..905906d01 100644
--- a/code_review_graph/skills.py
+++ b/code_review_graph/skills.py
@@ -680,20 +680,20 @@ def _record_configured(key: str, plat: dict[str, Any]) -> None:
             "## Explore Codebase\n\n"
             "Use the code-review-graph MCP tools to explore and understand the codebase.\n\n"
             "### Steps\n\n"
-            "1. Run `list_graph_stats` to see overall codebase metrics.\n"
+            "1. Run `list_graph_stats_tool` to see overall codebase metrics.\n"
             "2. Run `get_architecture_overview_tool` for high-level community structure.\n"
-            "3. Use `list_communities_tool` to find major modules, then `get_community` "
+            "3. Use `list_communities_tool` to find major modules, then `get_community_tool` "
             "for details.\n"
             "4. Use `semantic_search_nodes_tool` to find specific functions or classes.\n"
             "5. Use `query_graph_tool` with patterns like `callers_of`, `callees_of`, "
             "`imports_of` to trace relationships.\n"
-            "6. Use `list_flows` and `get_flow` to understand execution paths.\n\n"
+            "6. Use `list_flows_tool` and `get_flow_tool` to understand execution paths.\n\n"
             "### Tips\n\n"
             "- Start broad (stats, architecture) then narrow down to specific areas.\n"
             "- Use `children_of` on a file to see all its functions and classes.\n"
-            "- Use `find_large_functions` to identify complex code.\n\n"
+            "- Use `find_large_functions_tool` to identify complex code.\n\n"
             "## Token Efficiency Rules\n"
-            '- ALWAYS start with `get_minimal_context(task="<your task>")` '
+            '- ALWAYS start with `get_minimal_context_tool(task="<your task>")` '
             "before any other graph tool.\n"
             '- Use `detail_level="minimal"` on all calls. Only escalate to '
             '"standard" when minimal is insufficient.\n'
@@ -721,7 +721,7 @@ def _record_configured(key: str, plat: dict[str, Any]) -> None:
             "- Suggested improvements\n"
             "- Overall merge recommendation\n\n"
             "## Token Efficiency Rules\n"
-            '- ALWAYS start with `get_minimal_context(task="<your task>")` '
+            '- ALWAYS start with `get_minimal_context_tool(task="<your task>")` '
             "before any other graph tool.\n"
             '- Use `detail_level="minimal"` on all calls. Only escalate to '
             '"standard" when minimal is insufficient.\n'
@@ -739,7 +739,7 @@ def _record_configured(key: str, plat: dict[str, Any]) -> None:
             "1. Use `semantic_search_nodes_tool` to find code related to the issue.\n"
             "2. Use `query_graph_tool` with `callers_of` and `callees_of` to trace "
             "call chains.\n"
-            "3. Use `get_flow` to see full execution paths through suspected areas.\n"
+            "3. Use `get_flow_tool` to see full execution paths through suspected areas.\n"
             "4. Run `detect_changes_tool` to check if recent changes caused the issue.\n"
             "5. Use `get_impact_radius_tool` on suspected files to see what else is affected.\n\n"
             "### Tips\n\n"
@@ -747,7 +747,7 @@ def _record_configured(key: str, plat: dict[str, Any]) -> None:
             "- Look at affected flows to find the entry point that triggers the bug.\n"
             "- Recent changes are the most common source of new issues.\n\n"
             "## Token Efficiency Rules\n"
-            '- ALWAYS start with `get_minimal_context(task="<your task>")` '
+            '- ALWAYS start with `get_minimal_context_tool(task="<your task>")` '
             "before any other graph tool.\n"
             '- Use `detail_level="minimal"` on all calls. Only escalate to '
             '"standard" when minimal is insufficient.\n'
@@ -773,9 +773,9 @@ def _record_configured(key: str, plat: dict[str, Any]) -> None:
             "- Always preview before applying (rename mode gives you an edit list).\n"
             "- Check `get_impact_radius_tool` before major refactors.\n"
             "- Use `get_affected_flows_tool` to ensure no critical paths are broken.\n"
-            "- Run `find_large_functions` to identify decomposition targets.\n\n"
+            "- Run `find_large_functions_tool` to identify decomposition targets.\n\n"
             "## Token Efficiency Rules\n"
-            '- ALWAYS start with `get_minimal_context(task="<your task>")` '
+            '- ALWAYS start with `get_minimal_context_tool(task="<your task>")` '
             "before any other graph tool.\n"
             '- Use `detail_level="minimal"` on all calls. Only escalate to '
             '"standard" when minimal is insufficient.\n'
diff --git a/code_review_graph/tools/community_tools.py b/code_review_graph/tools/community_tools.py
index 4c6647c65..eb2334e9f 100644
--- a/code_review_graph/tools/community_tools.py
+++ b/code_review_graph/tools/community_tools.py
@@ -56,7 +56,7 @@ def list_communities_func(
             "communities": communities,
         }
         result["_hints"] = generate_hints(
-            "list_communities", result, get_session()
+            "list_communities_tool", result, get_session()
         )
         return result
     except Exception as exc:
@@ -132,7 +132,7 @@ def get_community_func(
             "community": community,
         }
         result["_hints"] = generate_hints(
-            "get_community", result, get_session()
+            "get_community_tool", result, get_session()
         )
         return result
     except Exception as exc:
@@ -235,7 +235,7 @@ def get_architecture_overview_func(
             **overview,
         }
         result["_hints"] = generate_hints(
-            "get_architecture_overview", result, get_session()
+            "get_architecture_overview_tool", result, get_session()
         )
         if detail_level == "minimal":
             attach_context_savings(result, original_context=full_overview)
diff --git a/code_review_graph/tools/flows_tools.py b/code_review_graph/tools/flows_tools.py
index 4df1f713c..073ca7c8b 100644
--- a/code_review_graph/tools/flows_tools.py
+++ b/code_review_graph/tools/flows_tools.py
@@ -73,7 +73,7 @@ def list_flows(
             "flows": flows,
         }
         result["_hints"] = generate_hints(
-            "list_flows", result, get_session()
+            "list_flows_tool", result, get_session()
         )
         return result
     except Exception as exc:
@@ -167,7 +167,7 @@ def get_flow(
             "flow": flow,
         }
         result["_hints"] = generate_hints(
-            "get_flow", result, get_session()
+            "get_flow_tool", result, get_session()
         )
         return result
     except Exception as exc:
diff --git a/code_review_graph/tools/query.py b/code_review_graph/tools/query.py
index 5fdbd1e36..fb31735db 100644
--- a/code_review_graph/tools/query.py
+++ b/code_review_graph/tools/query.py
@@ -765,7 +765,7 @@ def semantic_search_nodes(
             "results": results,
         }
         result["_hints"] = generate_hints(
-            "semantic_search_nodes", result, get_session()
+            "semantic_search_nodes_tool", result, get_session()
         )
         return result
     finally:
diff --git a/code_review_graph/tools/refactor_tools.py b/code_review_graph/tools/refactor_tools.py
index c92891b3f..fe03efeff 100644
--- a/code_review_graph/tools/refactor_tools.py
+++ b/code_review_graph/tools/refactor_tools.py
@@ -84,7 +84,7 @@ def refactor_func(
                 **preview,
             }
             result["_hints"] = generate_hints(
-                "refactor", result, get_session()
+                "refactor_tool", result, get_session()
             )
             return result
 
@@ -99,7 +99,7 @@ def refactor_func(
                 "total": len(dead),
             }
             result["_hints"] = generate_hints(
-                "refactor", result, get_session()
+                "refactor_tool", result, get_session()
             )
             return result
 
@@ -115,7 +115,7 @@ def refactor_func(
                 "total": len(suggestions),
             }
             result["_hints"] = generate_hints(
-                "refactor", result, get_session()
+                "refactor_tool", result, get_session()
             )
             return result
 
@@ -165,4 +165,7 @@ def apply_refactor_func(
         return {"status": "error", "error": str(exc)}
 
     result = apply_refactor(refactor_id, root, dry_run=dry_run)
+    result["_hints"] = generate_hints(
+        "apply_refactor_tool", result, get_session()
+    )
     return result
diff --git a/code_review_graph/tools/review.py b/code_review_graph/tools/review.py
index 9b8775922..44de38b0e 100644
--- a/code_review_graph/tools/review.py
+++ b/code_review_graph/tools/review.py
@@ -114,8 +114,8 @@ def get_review_context(
                 "key_entities": key_entities,
                 "test_gaps": test_gap_count,
                 "next_tool_suggestions": [
-                    "detect_changes",
-                    "get_affected_flows",
+                    "detect_changes_tool",
+                    "get_affected_flows_tool",
                     "get_impact_radius",
                 ],
             }
@@ -341,7 +341,7 @@ def get_affected_flows_func(
             "total": total,
         }
         out["_hints"] = generate_hints(
-            "get_affected_flows", out, get_session()
+            "get_affected_flows_tool", out, get_session()
         )
         return out
     except Exception as exc:
@@ -470,7 +470,7 @@ def detect_changes_func(
                 **analysis,
             }
         result["_hints"] = generate_hints(
-            "detect_changes", result, get_session()
+            "detect_changes_tool", result, get_session()
         )
         attach_context_savings(result, original_tokens=original_tokens)
         return result
diff --git a/skills/debug-issue/SKILL.md b/skills/debug-issue/SKILL.md
index b8d928fa3..a0fc76935 100644
--- a/skills/debug-issue/SKILL.md
+++ b/skills/debug-issue/SKILL.md
@@ -11,7 +11,7 @@ Use the knowledge graph to systematically trace and debug issues.
 
 1. Use `semantic_search_nodes_tool` to find code related to the issue.
 2. Use `query_graph_tool` with `callers_of` and `callees_of` to trace call chains.
-3. Use `get_flow` to see full execution paths through suspected areas.
+3. Use `get_flow_tool` to see full execution paths through suspected areas.
 4. Run `detect_changes_tool` to check if recent changes caused the issue.
 5. Use `get_impact_radius_tool` on suspected files to see what else is affected.
 
@@ -22,6 +22,6 @@ Use the knowledge graph to systematically trace and debug issues.
 - Recent changes are the most common source of new issues.
 
 ## Token Efficiency Rules
-- ALWAYS start with `get_minimal_context(task="<your task>")` before any other graph tool.
+- ALWAYS start with `get_minimal_context_tool(task="<your task>")` before any other graph tool.
 - Use `detail_level="minimal"` on all calls. Only escalate to "standard" when minimal is insufficient.
 - Target: complete any review/debug/refactor task in ≤5 tool calls and ≤800 total output tokens.
diff --git a/skills/explore-codebase/SKILL.md b/skills/explore-codebase/SKILL.md
index a470b3c8d..74dec19b2 100644
--- a/skills/explore-codebase/SKILL.md
+++ b/skills/explore-codebase/SKILL.md
@@ -9,20 +9,20 @@ Use the code-review-graph MCP tools to explore and understand the codebase.
 
 ### Steps
 
-1. Run `list_graph_stats` to see overall codebase metrics.
+1. Run `list_graph_stats_tool` to see overall codebase metrics.
 2. Run `get_architecture_overview_tool` for high-level community structure.
-3. Use `list_communities_tool` to find major modules, then `get_community` for details.
+3. Use `list_communities_tool` to find major modules, then `get_community_tool` for details.
 4. Use `semantic_search_nodes_tool` to find specific functions or classes.
 5. Use `query_graph_tool` with patterns like `callers_of`, `callees_of`, `imports_of` to trace relationships.
-6. Use `list_flows` and `get_flow` to understand execution paths.
+6. Use `list_flows_tool` and `get_flow_tool` to understand execution paths.
 
 ### Tips
 
 - Start broad (stats, architecture) then narrow down to specific areas.
 - Use `children_of` on a file to see all its functions and classes.
-- Use `find_large_functions` to identify complex code.
+- Use `find_large_functions_tool` to identify complex code.
 
 ## Token Efficiency Rules
-- ALWAYS start with `get_minimal_context(task="<your task>")` before any other graph tool.
+- ALWAYS start with `get_minimal_context_tool(task="<your task>")` before any other graph tool.
 - Use `detail_level="minimal"` on all calls. Only escalate to "standard" when minimal is insufficient.
 - Target: complete any review/debug/refactor task in ≤5 tool calls and ≤800 total output tokens.
diff --git a/skills/refactor-safely/SKILL.md b/skills/refactor-safely/SKILL.md
index 3020efb5c..410b5c397 100644
--- a/skills/refactor-safely/SKILL.md
+++ b/skills/refactor-safely/SKILL.md
@@ -20,9 +20,9 @@ Use the knowledge graph to plan and execute refactoring with confidence.
 - Always preview before applying (rename mode gives you an edit list).
 - Check `get_impact_radius_tool` before major refactors.
 - Use `get_affected_flows_tool` to ensure no critical paths are broken.
-- Run `find_large_functions` to identify decomposition targets.
+- Run `find_large_functions_tool` to identify decomposition targets.
 
 ## Token Efficiency Rules
-- ALWAYS start with `get_minimal_context(task="<your task>")` before any other graph tool.
+- ALWAYS start with `get_minimal_context_tool(task="<your task>")` before any other graph tool.
 - Use `detail_level="minimal"` on all calls. Only escalate to "standard" when minimal is insufficient.
 - Target: complete any review/debug/refactor task in ≤5 tool calls and ≤800 total output tokens.
diff --git a/skills/review-changes/SKILL.md b/skills/review-changes/SKILL.md
index a5c5e68cd..c62810d2b 100644
--- a/skills/review-changes/SKILL.md
+++ b/skills/review-changes/SKILL.md
@@ -24,6 +24,6 @@ Provide findings grouped by risk level (high/medium/low) with:
 - Overall merge recommendation
 
 ## Token Efficiency Rules
-- ALWAYS start with `get_minimal_context(task="<your task>")` before any other graph tool.
+- ALWAYS start with `get_minimal_context_tool(task="<your task>")` before any other graph tool.
 - Use `detail_level="minimal"` on all calls. Only escalate to "standard" when minimal is insufficient.
 - Target: complete any review/debug/refactor task in ≤5 tool calls and ≤800 total output tokens.
diff --git a/tests/test_hints.py b/tests/test_hints.py
index 14d9967d3..f4ac82f19 100644
--- a/tests/test_hints.py
+++ b/tests/test_hints.py
@@ -19,24 +19,30 @@ def test_fresh_session_exploring(self):
     def test_review_intent_detected(self):
         """Recording review-oriented tools should infer 'reviewing'."""
         session = SessionState()
-        for tool in ("detect_changes", "get_review_context", "get_affected_flows"):
+        for tool in ("detect_changes_tool", "get_review_context_tool", "get_affected_flows_tool"):
             session.record_tool_call(tool)
         assert infer_intent(session) == "reviewing"
 
     def test_debug_intent_detected(self):
         """Recording debug-oriented tools should infer 'debugging'."""
         session = SessionState()
-        for tool in ("query_graph", "get_flow", "semantic_search_nodes"):
+        for tool in ("query_graph_tool", "get_flow_tool", "semantic_search_nodes_tool"):
             session.record_tool_call(tool)
         assert infer_intent(session) == "debugging"
 
     def test_refactoring_intent_detected(self):
         """Recording refactoring-oriented tools should infer 'refactoring'."""
         session = SessionState()
-        for tool in ("refactor", "find_dead_code", "suggest_refactorings"):
+        for tool in ("refactor_tool", "apply_refactor_tool"):
             session.record_tool_call(tool)
         assert infer_intent(session) == "refactoring"
 
+    def test_large_functions_tool_does_not_force_refactoring_intent(self):
+        """find_large_functions_tool is a size query, not a refactor action."""
+        session = SessionState()
+        session.record_tool_call("find_large_functions_tool")
+        assert infer_intent(session) == "exploring"
+
     def test_session_caps_history(self):
         """tools_called should never exceed 100 entries (FIFO)."""
         session = SessionState()
@@ -58,28 +64,28 @@ class TestGenerateHints:
     def test_hints_no_repeat(self):
         """Already-called tools must not appear in next_steps."""
         session = SessionState()
-        # Call list_flows, then generate hints for it
-        # list_flows suggests get_flow, get_affected_flows, get_architecture_overview
-        generate_hints("list_flows", {"status": "ok"}, session)
+        # Call list_flows_tool, then generate hints for it
+        # list_flows_tool suggests get_flow_tool, get_affected_flows_tool, get_architecture_overview_tool
+        generate_hints("list_flows_tool", {"status": "ok"}, session)
 
-        # Now call get_flow and regenerate hints for list_flows
-        hints2 = generate_hints("list_flows", {"status": "ok"}, session)
+        # Now call get_flow_tool and regenerate hints for list_flows_tool
+        hints2 = generate_hints("list_flows_tool", {"status": "ok"}, session)
         suggested_tools2 = {s["tool"] for s in hints2["next_steps"]}
-        # list_flows itself was called, so it shouldn't be suggested by get_flow workflow
-        # Also, the first list_flows call should be excluded from next suggestions
-        assert "list_flows" not in suggested_tools2
+        # list_flows_tool itself was called, so it should not be suggested by the get_flow_tool workflow
+        # Also, the first list_flows_tool call should be excluded from next suggestions
+        assert "list_flows_tool" not in suggested_tools2
 
     def test_hints_max_three(self):
         """Each hints category should have at most 3 entries."""
         session = SessionState()
-        # detect_changes has 4 workflow entries
+        # detect_changes_tool has 4 workflow entries
         result = {
             "status": "ok",
             "test_gaps": [{"name": f"gap_{i}"} for i in range(10)],
             "risk_score": 0.9,
             "warnings": ["coupling warning 1", "coupling warning 2"],
         }
-        hints = generate_hints("detect_changes", result, session)
+        hints = generate_hints("detect_changes_tool", result, session)
         assert len(hints["next_steps"]) <= _MAX_PER_CATEGORY
         assert len(hints["warnings"]) <= _MAX_PER_CATEGORY
         assert len(hints["related"]) <= _MAX_PER_CATEGORY
@@ -91,7 +97,7 @@ def test_warnings_from_result_test_gaps(self):
             "status": "ok",
             "test_gaps": [{"name": "untested_func"}, {"name": "another_func"}],
         }
-        hints = generate_hints("detect_changes", result, session)
+        hints = generate_hints("detect_changes_tool", result, session)
         assert any("Test coverage gaps" in w for w in hints["warnings"])
         assert any("untested_func" in w for w in hints["warnings"])
 
@@ -99,20 +105,20 @@ def test_warnings_from_result_risk_score(self):
         """High risk_score in result should produce a warning."""
         session = SessionState()
         result = {"status": "ok", "risk_score": 0.85}
-        hints = generate_hints("detect_changes", result, session)
+        hints = generate_hints("detect_changes_tool", result, session)
         assert any("High risk score" in w for w in hints["warnings"])
 
     def test_warnings_low_risk_no_warning(self):
         """Low risk_score should NOT produce a warning."""
         session = SessionState()
         result = {"status": "ok", "risk_score": 0.3}
-        hints = generate_hints("detect_changes", result, session)
+        hints = generate_hints("detect_changes_tool", result, session)
         assert not any("High risk score" in w for w in hints["warnings"])
 
     def test_generate_hints_empty_result(self):
         """An empty/minimal result should still return valid hints structure."""
         session = SessionState()
-        hints = generate_hints("list_flows", {}, session)
+        hints = generate_hints("list_flows_tool", {}, session)
         assert "next_steps" in hints
         assert "related" in hints
         assert "warnings" in hints
@@ -131,7 +137,7 @@ def test_session_records_files(self):
         """Files from result should be tracked in session state."""
         session = SessionState()
         result = {"status": "ok", "changed_files": ["a.py", "b.py"]}
-        generate_hints("detect_changes", result, session)
+        generate_hints("detect_changes_tool", result, session)
         assert "a.py" in session.files_touched
         assert "b.py" in session.files_touched
 
@@ -145,7 +151,7 @@ def test_session_records_nodes(self):
                 {"qualified_name": "mod.py::Bar", "name": "Bar"},
             ],
         }
-        generate_hints("semantic_search_nodes", result, session)
+        generate_hints("semantic_search_nodes_tool", result, session)
         assert "mod.py::Foo" in session.nodes_queried
         assert "mod.py::Bar" in session.nodes_queried
 
@@ -157,10 +163,35 @@ def test_related_suggests_untouched_files(self):
             "status": "ok",
             "impacted_files": ["already_seen.py", "new_file.py", "other.py"],
         }
-        hints = generate_hints("detect_changes", result, session)
+        hints = generate_hints("detect_changes_tool", result, session)
         assert "already_seen.py" not in hints["related"]
         assert "new_file.py" in hints["related"]
 
+    def test_apply_refactor_hints(self):
+        """apply_refactor_tool should surface follow-up review actions."""
+        session = SessionState()
+        result = {
+            "status": "ok",
+            "applied": True,
+            "changed_files": ["a.py"],
+        }
+        hints = generate_hints("apply_refactor_tool", result, session)
+        suggested_tools = {step["tool"] for step in hints["next_steps"]}
+        assert "detect_changes_tool" in suggested_tools
+        assert "get_affected_flows_tool" in suggested_tools
+
+    def test_refactor_hints_include_related_symbol_search(self):
+        """refactor_tool should suggest searching related symbols to rename."""
+        session = SessionState()
+        result = {
+            "status": "ok",
+            "previewed": True,
+            "changes": [],
+        }
+        hints = generate_hints("refactor_tool", result, session)
+        suggested_tools = {step["tool"] for step in hints["next_steps"]}
+        assert "semantic_search_nodes_tool" in suggested_tools
+
 
 class TestGlobalSession:
     def test_get_session_returns_singleton(self):
@@ -190,6 +221,6 @@ def test_warnings_from_arch_overview_dict(self):
                 {"message": "Circular dependency detected"},
             ],
         }
-        hints = generate_hints("get_architecture_overview", result, session)
+        hints = generate_hints("get_architecture_overview_tool", result, session)
         assert any("High coupling" in w for w in hints["warnings"])
         assert any("Circular dependency" in w for w in hints["warnings"])
diff --git a/tests/test_prompts.py b/tests/test_prompts.py
index 1714342f6..fa6a069e4 100644
--- a/tests/test_prompts.py
+++ b/tests/test_prompts.py
@@ -39,11 +39,11 @@ def test_custom_base(self):
 
     def test_mentions_detect_changes(self):
         result = review_changes_prompt()
-        assert "detect_changes" in _text(result[0])
+        assert "detect_changes_tool" in _text(result[0])
 
     def test_mentions_affected_flows(self):
         result = review_changes_prompt()
-        assert "affected_flows" in _text(result[0])
+        assert "get_affected_flows_tool" in _text(result[0])
 
     def test_mentions_test_gaps(self):
         result = review_changes_prompt()
@@ -96,11 +96,11 @@ def test_empty_description(self):
 
     def test_mentions_search(self):
         result = debug_issue_prompt(description="test issue")
-        assert "semantic_search_nodes" in _text(result[0])
+        assert "semantic_search_nodes_tool" in _text(result[0])
 
     def test_mentions_get_minimal_context(self):
         result = debug_issue_prompt()
-        assert "get_minimal_context" in _text(result[0])
+        assert "get_minimal_context_tool" in _text(result[0])
 
 
 class TestOnboardDeveloperPrompt:
@@ -118,7 +118,7 @@ def test_message_has_role_and_content(self):
 
     def test_mentions_stats(self):
         result = onboard_developer_prompt()
-        assert "list_graph_stats" in _text(result[0])
+        assert "list_graph_stats_tool" in _text(result[0])
 
     def test_mentions_architecture(self):
         result = onboard_developer_prompt()
@@ -158,13 +158,21 @@ def test_mentions_risk_scoring(self):
         result = pre_merge_check_prompt()
         assert "risk" in _text(result[0]).lower()
 
+    def test_mentions_related_symbol_search(self):
+        result = pre_merge_check_prompt()
+        assert "semantic_search_nodes_tool" in _text(result[0])
+
     def test_mentions_test_gaps(self):
         result = pre_merge_check_prompt()
-        assert "tests_for" in _text(result[0])
+        assert "query_graph_tool" in _text(result[0])
+
+    def test_mentions_dead_code_mode(self):
+        result = pre_merge_check_prompt()
+        assert 'refactor_tool(mode="dead_code"' in _text(result[0])
 
-    def test_mentions_dead_code(self):
+    def test_mentions_large_functions_tool(self):
         result = pre_merge_check_prompt()
-        assert "dead_code" in _text(result[0])
+        assert "find_large_functions_tool" in _text(result[0])
 
 
 class TestTokenEfficiencyPreamble:
@@ -172,21 +180,21 @@ class TestTokenEfficiencyPreamble:
 
     def test_review_has_preamble(self):
         result = review_changes_prompt()
-        assert "get_minimal_context" in _text(result[0])
+        assert "get_minimal_context_tool" in _text(result[0])
         assert "detail_level" in _text(result[0])
 
     def test_architecture_has_preamble(self):
         result = architecture_map_prompt()
-        assert "get_minimal_context" in _text(result[0])
+        assert "get_minimal_context_tool" in _text(result[0])
 
     def test_debug_has_preamble(self):
         result = debug_issue_prompt()
-        assert "get_minimal_context" in _text(result[0])
+        assert "get_minimal_context_tool" in _text(result[0])
 
     def test_onboard_has_preamble(self):
         result = onboard_developer_prompt()
-        assert "get_minimal_context" in _text(result[0])
+        assert "get_minimal_context_tool" in _text(result[0])
 
     def test_pre_merge_has_preamble(self):
         result = pre_merge_check_prompt()
-        assert "get_minimal_context" in _text(result[0])
+        assert "get_minimal_context_tool" in _text(result[0])
diff --git a/tests/test_skills.py b/tests/test_skills.py
index dd54eca90..5aac357d7 100644
--- a/tests/test_skills.py
+++ b/tests/test_skills.py
@@ -170,14 +170,51 @@ def test_custom_skills_dir(self, tmp_path):
         assert len(list(result.iterdir())) == 4
 
     def test_skill_content_includes_get_minimal_context(self, tmp_path):
-        """Every skill template must reference get_minimal_context."""
+        """Every skill template must reference get_minimal_context_tool."""
         skills_dir = generate_skills(tmp_path)
         for subdir in skills_dir.iterdir():
             content = (subdir / "SKILL.md").read_text()
-            assert "get_minimal_context" in content, (
-                f"{subdir.name} missing get_minimal_context reference"
+            assert "get_minimal_context_tool" in content, (
+                f"{subdir.name} missing get_minimal_context_tool reference"
             )
 
+    def test_skill_templates_use_exported_tool_names(self, tmp_path):
+        generated = generate_skills(tmp_path)
+        bundled = Path(__file__).parents[1] / "skills"
+
+        expected_tools = {
+            "explore-codebase": [
+                "get_minimal_context_tool",
+                "list_graph_stats_tool",
+                "get_community_tool",
+                "list_flows_tool",
+                "get_flow_tool",
+                "find_large_functions_tool",
+            ],
+            "review-changes": ["get_minimal_context_tool", "get_affected_flows_tool"],
+            "debug-issue": ["get_minimal_context_tool", "get_flow_tool"],
+            "refactor-safely": ["get_minimal_context_tool", "find_large_functions_tool"],
+        }
+        legacy_tools = [
+            "get_minimal_context",
+            "list_graph_stats",
+            "get_community",
+            "list_flows",
+            "get_flow",
+            "find_large_functions",
+        ]
+
+        for skill_name, tool_names in expected_tools.items():
+            for skill_file in (
+                generated / skill_name / "SKILL.md",
+                bundled / skill_name / "SKILL.md",
+            ):
+                content = skill_file.read_text(encoding="utf-8")
+                for tool_name in tool_names:
+                    assert tool_name in content, skill_file
+                for legacy_tool in legacy_tools:
+                    assert f"`{legacy_tool}`" not in content, skill_file
+
     def test_skill_content_includes_detail_level(self, tmp_path):
         """Every skill template must reference detail_level."""
         skills_dir = generate_skills(tmp_path)