i've run 4,000+ claude code sessions. this is everything i learned — a plugin, hooks, agents, and my actual takes on what works.
this repo maintains itself using the patterns it teaches.
claude plugin add anipotts/claude-code-tipsyou get mine — session mining to sqlite. costs, search, error memory, pattern detection. data stays local at ~/.claude/mine.db.
/mine → today's sessions, cost, top tools
/mine search "websocket" → full-text search across all conversations
/mine mistakes → error patterns claude keeps repeating
/mine hotspots → most-edited files across sessions
start with mine + safety-guard hook. add more as you go. mine docs →
copy one, wire it up, done. see hooks reference →
| hook | event | what it does |
|---|---|---|
| safety-guard | PreToolUse | blocks force push, rm -rf /, DROP TABLE |
| no-squash | PreToolUse | blocks squash merges |
| panopticon | PostToolUse | logs every tool call to sqlite |
| context-save | PreCompact | saves context before compression |
| notify | Notification | routes to macOS, Slack, ntfy |
| commit-nudge | PostToolUse | reminds you to commit after N edits |
| replay-capture | PostToolUse | captures file changes for VHS replays |
| version-stamp | SessionEnd | auto-updates tested-with stamps |
| stale-branch | SessionStart | warns about gone tracking branches |
| md-lint-fix | PostToolUse | auto-fixes markdown lint on save |
| knowledge-builder | PostToolUse | builds a codebase knowledge graph |
| agent | what it does |
|---|---|
| explorer | parallel worktree exploration — try risky changes safely |
| analyst | free-form SQL against mine.db |
| test-writer | generates edge case tests you missed |
| guardian | watches your project, proposes fixes |
| code-sweeper | finds dead code, unused imports |
| pr-narrator | writes PR descriptions from your diff |
| vibe-check | quick architecture review |
| command | what it does |
|---|---|
/mine |
usage data — costs, sessions, search |
/ship |
stage, commit, push, open PR |
/sweep |
clean dead code, unused imports |
/quicktest |
find and run tests for current file |
/deps |
dependency updates and security audit |
/improve |
propose CLAUDE.md updates from git history |
| what | who it's for | |
|---|---|---|
| concepts | subagent patterns, automation, cost optimization | anyone using agentic coding tools |
| claude code | guide, hooks, plugins, MCP, CLI | claude code users |
| comparisons | vs cursor, vs codex, vs gemini, pricing | choosing a tool |
| my tips | why hooks matter, my stack, mistakes i made, cost reality | my personal takes |
concepts/ is tool-agnostic. tips/ is me.
this repo is its own best example. what's running →
- 8 CI pipelines — upstream watcher, competitive intel, community digest, docs audit, freshness, stale cleanup, dependabot, releases
- 11 hooks, 10 agents, 8 commands — all running on this repo
- 0 manual maintenance — everything that doesn't require taste is automated
- CLAUDE.md templates — starter configs for TypeScript, Python, Rust, Next.js
- handoff plugin — PreCompact context preservation
- broadcast plugin — async notifications on git events
MIT · built by anipotts from 4,000+ sessions
