By eigenone
Your AI doesn't remember. Jeeves does. Session memory and living documentation for Claude Code.
Activate Jeeves with your API key. Use when user says "activate", provides a jvs_ key, or after signing up at trustjeeves.com.
Find complex code with few comments, add WHY comments. Use when user says "annotate", "add comments to code".
Stash current thinking, start fresh. Use when user says "start fresh", "archive this", "clean slate".
Plan what docs to create — analyzes code directories and suggests missing pattern/decision docs. Use when user says "what docs should I create?", "plan the knowledge base".
Compare specs and plans against what was actually built. Use when user says "drift check", "compare spec vs code", "what did we skip?".
Admin access level
Server config contains admin-level keywords
Executes bash commands
Hook triggers when Bash tool is used
Own this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimOwn this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimBased on adoption, maintenance, documentation, and repository signals. Not a security audit or endorsement.
Your AI doesn't remember. Jeeves does.
Session memory and living documentation for Claude Code.
/plugin marketplace add eigenone/jeeves
/plugin install jeeves@eigenone-jeeves
Then reload: /reload-plugins
curl -X POST https://jeeves-api.singhal-priyank.workers.dev/signup -H "Content-Type: application/json" -d '{"email":"[email protected]","persona":"builder"}'export JEEVES_KEY=jvs_your_key_here (add to .zshrc)| Command | What it does |
|---|---|
/jeeves | Sync docs with code — heal paths, rebuild index, report health |
/jeeves:end | End session — capture decisions, write handoff |
/jeeves:summary | Everything decided across all sessions |
/jeeves:harden | Quality suite — annotate code, verify comments, lint |
/jeeves:research | Web research, save findings |
/jeeves:export | Shareable doc for your team |
/jeeves:trace | End-to-end feature flow |
/jeeves:design | Plan what docs to create |
/jeeves:reconcile | Check docs for drift |
/jeeves:driftcheck | Compare specs vs actual code |
/jeeves:annotate | Add WHY comments to code |
/jeeves:verify | Check comment accuracy |
/jeeves:save | Save artifacts (drafts, plans) |
/jeeves:archive | Stash thinking, start fresh |
/jeeves:extract | File back knowledge from conversation |
/plugin update jeeves@eigenone-jeeves
Then restart Claude Code — a running session keeps the old plugin code loaded until it reloads. Jeeves prints a one-line warning when the version on disk is newer than the session loaded.
If your repo has its own scripts/ copies of Jeeves (run git ls-files | grep scripts/heal-docs.ts — if it prints a path, you do), note that a plugin update only refreshes the copy inside the plugin, not files committed to your repo. The slash commands always use the upgraded plugin copy, but the auto-heal step on a /jeeves sync prefers your local scripts/heal-docs.ts when present. After updating, either rm scripts/heal-docs.ts (recommended — it then falls back to the plugin copy) or re-copy the script in so your local version matches. Updates change future behavior only; they don't revert edits an older version already wrote.
Free (forever): /jeeves, /jeeves:end, /jeeves:summary
Pro (14-day trial): All 15 commands
https://jeeves-api.singhal-priyank.workers.devnpx claudepluginhub eigenone/jeeves --plugin jeevesUpstash Context7 MCP server for up-to-date documentation lookup. Pull version-specific documentation and code examples directly from source repositories into your LLM context.
Permanent coding companion for Claude Code — survives any update. MCP-based terminal pet with ASCII art, stats, reactions, and personality.
Upstash Context7 MCP server for up-to-date documentation lookup. Pull version-specific documentation and code examples directly from source repositories into your LLM context.
Build and maintain an LLM-curated personal knowledge base in your project — Andrej Karpathy's LLM Wiki pattern, designed to scale to thousands of pages without becoming a context bottleneck. Now with an optional compiled graph layer for typed, provenance-backed relationships.
Comprehensive C4 architecture documentation workflow with bottom-up code analysis, component synthesis, container mapping, and context diagram generation