Add MseeP.ai badge#7
Open
mseep-ai wants to merge 175 commits intocentminmod:masterfrom
Open
Conversation
Add MIT license
…educe reasoning tokens with minimal loss in context
- enable thinking by default with MAX_THINKING_TOKENS = 8192. Results in better responses - remove allowed permission for rm deleting files by default
…ssion-report Adds three new auto-hiding sections at every scope (session / --project-cost / --all-projects), inspired by Anthropic's official session-report skill: - Cache breaks: single turns whose input + cache_creation exceed a configurable threshold (--cache-break-threshold, default 100_000), each with ±2 user-prompt context slice. - Skills & slash commands: invocations / turns / input / cache % / output / cost per named skill or /slash command. Slash prefixes de-slashed so they merge with Skill-tool invocations of the same name. - Subagent types: spawn count always; token + cost under --include- subagents, with meta.json → filename-regex → fork fallback for type. Also adds UUID-based cross-file dedup at project + instance scope so resumed sessions can't double-count; single-session scope keeps the message.id streaming-split dedup. HTML/MD/CSV/JSON coverage. 14 new unit tests, 468 total passing.
…t attribution Rolls subagent token usage onto the user prompt that spawned the chain via three new additive turn-record fields (attributed_subagent_*). HTML prompts table sorts by parent + attributed cost by default and shows a "Subagents +$" column when any top-N row has attribution. Includes nested-chain support, cycle guard, two new CLI flags (--no-subagent-attribution, --sort-prompts-by), 14 new unit tests, and documentation updates.
…ss all 4 theme variants
Fix console theme turn drawer showing transparent background — the group panel selector set background to rgba(165,139,255,.03) on all panels including the drawer; override restores var(--bg) (#08080A).
Adds per-turn latency + session wall-clock surfacing, the new benchmark-effort-prompt multi-variant sweep skill, latency charts for both benchmark skills, and bugfixes from the dual-AI code review (stale timestamp inheritance, cross-side averaging halving, 8-col fallback row, heatmap all-NaN guard, list-content scanner depth).
…cation Adds Turn Character & Efficiency Signals: classifies every assistant turn into one of 9 waste categories (productive, retry, cache-heavy, cache payload, reasoning, file re-read, verbose edit, stuck/truncated, subagent dispatch). Dashboard shows stacked bar + drill-down cards; detail timeline shows Turn Character column; JSON/CSV/MD exports include new waste fields.
…ross-browser overflow fix (v1.23.0)
…cess, allowlist BASH_PATH_RE (v1.24.0)
… Advisor calls card
When a turn carries advisor calls, the per-turn drawer now shows Advisor input and Advisor output token rows beneath Cache write in the TOKENS section. Rows are hidden on non-advisor turns.
Bug fix: guard against iterations:null in usage when advisor is not enabled. Synthetic resume-marker turns from environments without the advisor feature emit "iterations": null rather than omitting the key, causing TypeError on project-scope runs. Fixed with `or []` guard in _cost and _advisor_info.
… within-session split + warm-up columns - Headline "Subagent share of cost" KPI card with --include-subagents branching and lower-bound disclosure when orphan turns exist. Card structure mirrors the other KPI cards (kpi-label / big kpi-val / two thin kpi-sub lines). - Subagent attribution coverage section (themed models-table) — surfaces orphans, cycles, max nesting depth, spawn → attributed-turn fanout. - Within-session spawning split table — median combined cost on spawning vs. non-spawning turns per session (≥3 turns each bucket). Descriptive correlation, not a counterfactual estimate. - Two warm-up columns in by_subagent_type: median first-turn cost share + SP-amortisation rate. - Per-prompt badge: appended (NN% of combined cost). Labelled "combined" not "of turn". - Render-time only — no parser changes, no _SCRIPT_VERSION bump. - 8 new tests; 514 passed, 1 skipped.
…n + content-block undercount
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Hi there,
This pull request shares a security update on my-claude-code-setup.
We also have an entry for my-claude-code-setup in our directory, MseeP.ai, where we provide regular security and trust updates on your app.
We invite you to add our badge for your MCP server to your README to help your users learn from a third party that provides ongoing validation of my-claude-code-setup.
You can easily take control over your listing for free: visit it at https://mseep.ai/app/centminmod-my-claude-code-setup.
Thanks,
The MseeP Team
MCP servers you can trust