shepherdjerred/monorepo/history
Claude Code and agent skill by shepherdjerred/monorepo
- community
Checking that the SKILL.md file is still in the repository.
Search local Conductor, Claude Code, Codex, Cursor, Antigravity, and Grok work history with toolkit.
Install
npx vskill@latest install shepherdjerred/monorepo/historySKILL.md
Local agent history
Use the local index for recollection and prior-work discovery. It is not a deployment or CI status oracle.
First-time setup
The daemon is a user-scoped macOS LaunchAgent. It polls the source stores every 30 seconds and writes only to ~/.toolkit/history/. daemon reindex requests a full scan; daemon stop unloads the job, and daemon uninstall also removes the plist.
Search recipes
Supported sources are conductor, claude, codex, cursor, antigravity, and grok. Search uses BM25 relevance; recency only breaks ties. Unquoted terms are AND-prefix matches. To request an exact phrase, preserve literal quotes in the query, for example toolkit history search '"Match Reports"'.
Search and recent hide the current Conductor/Codex run and group parallel sessions that opened with the same normalized prompt. Use --include-current or --include-duplicates to override those defaults. --include-excerpts reopens only returned records and adds 360-character dialogue-first excerpts.
show is the normal second stage. By default it returns the opening request plus latest dialogue, bounded to eight messages and 6,000 characters. With --query, it centers on the best dialogue match and admits only matching tool messages; --include-tools admits all nearby tools. System instructions, reasoning, and compaction records are excluded. Cursor messages use unknown because its flattened index has no roles. IDs are local to the current rebuild; rerun search if an ID is missing.
Search JSON is { query, results, warnings }, recent JSON is { results, warnings }, and show JSON is { record, messages, truncated }. Human-readable source warnings go to stderr. An unavailable source is silent in an all-source query when it is merely uninstalled, but warns when explicitly requested.
For “what is the current status?”, use history to find the relevant branch, PR, or workspace, then verify the live state separately with toolkit deployed, toolkit pr health, or the relevant system client. Do not infer merged, deployed, reachable, or running state from a transcript.
Privacy and failure boundaries
- The index is a rebuildable contentless FTS5 database at
~/.toolkit/history/index.sqlite; the daemon socket, state, logs, and plist are user-private.
- Source SQLite databases and JSONL files are read-only. Schema changes are
reported per source through history sources; they are not silently ignored.
- Do not place transcript contents, index databases, daemon state, LaunchAgent
files, or credentials in the repository or .context/.
Security
- Security scanPassed, 95 of 100Pattern checks on the skill's files.
- Model reviewPassed
Quality evaluation
- Task evaluationNot runNo evaluation yet. Run one in Skill Studio.
Source
- SKILL.md in the repositoryCheckingpackages/toolkit/skills/history/SKILL.md
Popularity
- 7-day trend score
- 5
Not refreshed yet.
Works with
Any agent vskill installs to: 56 targets in all. These 39 have a page on this site.
- Claude Code
- Cursor
- GitHub Copilot
- Windsurf
- Codex
- Gemini CLI
- Cline
- Amp
- Roo Code
- Goose
- Continue
- Kilo Code
- Junie
- Trae
- OpenHands
- Qwen Code
- AI Agent
- Tabnine Chat
- Sourcegraph Cody
- Amazon Q
- Firebase Genkit
- Sweep
- Devin
- Phind
- Mentat
- Claude Desktop
- Aider
- GPT Pilot
- Plandex
- bolt.new
- v0
- Replit Agent
- Anse
- TypingMind
- BetterChatGPT
- LibreChat
- Warp AI
- Pieces
- Fig AI
Embed badge
[](https://verified-skill.com/skills/shepherdjerred/monorepo/history)