EN | ZH | ES | HI | PT | JA

claude-code-tips

my claude code setup, open source. hooks, agents, tips, and a plugin that mines your usage data.

if this saves you time, star it. it helps others find it.

quick start

/plugin marketplace add anipotts/claude-code-tips   # add marketplace (one time)
/plugin install mine@cc                             # install mine (session analytics)
/plugin install cc@cc                               # install cc (cross-session messaging)

then: copy safety-guard.sh to block dangerous commands. read a tip. done.

---

the numbers

hundreds of sessions across dozens of projects. $200/mo max plan.

same usage would cost ~$12K on the API with caching, ~$95K without. no autonomous loops. no cron jobs. every session starts with me typing a prompt. how the cost math works →

<img src="./gifs/mine-stats.gif" width="100%" alt="mine stats showing sessions, tokens, costs, and projects" />

---

install the mine plugin

/plugin marketplace add anipotts/claude-code-tips   # add marketplace (one time)
/plugin install mine@cc                             # install mine (session analytics)
/plugin install cc@cc                               # install cc (cross-session messaging)

you get mine · session mining to sqlite. costs, search, error memory, pattern detection. all 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
/mine loops               repeated patterns across sessions

start with mine + the safety-guard hook. add more as you go. mine docs &rarr;

---

cc plugin

cross-session messaging. see what other claude code sessions are doing, send messages between them.

/plugin install cc@cc

/cc                          show active sessions
/cc send merizo "pause"      message another session

---

the 3 things that changed how i code

hooks

hooks are the difference between "claude does what i want" and "claude does whatever it feels like." CLAUDE.md gives guidance. hooks give enforcement. one is a suggestion, the other is a wall.

this repo has 9 hooks you can drop into any project. safety-guard blocks force pushes, rm -rf /, and curl | bash. no-squash blocks squash merges. context-save preserves state before compaction. pick the ones that fit your workflow. hook guide &rarr;

agent teams

multiple claude instances working simultaneously on the same codebase, each in its own git worktree. the coordinator assigns tasks, collects results, merges the best approach.

i use this for parallel research, trying risky changes safely, and comparing approaches side-by-side without touching my working tree. how i use agent teams &rarr;

prompt caching

this is why the $200/mo plan is the best deal in AI coding. claude code caches your system prompt, tools, and CLAUDE.md as a prefix. 91% of my input tokens hit the cache, meaning i pay 10% of the input cost on 91% of my reads.

the key: keep your CLAUDE.md short and stable. every edit breaks the prefix cache. mine is 30 lines and changes maybe once a week. the full cost breakdown &rarr;

---

tips

short, standalone techniques. each one is something you can use in your next session.

tip	what you learn
prompt caching	get 97%+ cache hit rates, slash your bill
safety hooks	block force pushes and rm -rf in 5 minutes
settings hierarchy	project vs global vs local settings
session length	why shorter sessions are more efficient (with data)
ultrathink	force extended thinking for complex problems
context management	compaction strategies, active tool rate, keeping sessions tight
plan mode	when planning saves time vs when it wastes it
fast mode	same model, faster output, the tradeoff
plugins	build a plugin from scratch, what makes one worth installing
subagents	agent teams, worktree isolation, when parallel pays off
mcp integration	wire up MCP servers, use them inside sessions
hooks v2	command vs http vs prompt hooks, the async pattern

---

hooks

copy one, wire it up, done. each is a standalone bash script. full guide &rarr;

hook	event	what it does
safety-guard	PreToolUse	blocks force push, rm -rf /, DROP TABLE, curl-pipe-sh
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

<details> <summary>4 more hooks</summary>

hook	event	what it does
commit-nudge	PostToolUse	reminds you to commit after N edits
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

</details> <img src="./gifs/hook-safety.gif" width="100%" alt="safety-guard blocking a dangerous command" />

example agents

copy to .claude/agents/ and invoke with /agent <name>. each teaches a different pattern. guide &rarr;

agent	pattern	what it does
watch-tests	daemon	watches files, runs tests, proposes fixes
try-worktree	worktree	tries risky changes in isolated worktrees
arch-review	quick review	fast architecture smell-test
write-pr	git integration	PR descriptions from your diff

commands i use

command	what it does
/mine	usage data · costs, sessions, search, patterns
/ship	stage, commit, push, open PR in one command
/improve	propose CLAUDE.md updates from git history

plus 2 example commands you can copy: /sweep, /quicktest.

---

my personal takes

	what
cost reality	what claude code actually costs, the prompt caching math
mistakes i made	what burned me so you can skip it
automation	the 12 CI pipelines that maintain this repo
session workflow	how i work day-to-day with claude code
worktrees	parallel exploration with the desktop app

vs the alternatives

diplomatic, data-driven, no FUD. every claim cites a source.

vs cursor &middot; vs codex &middot; vs gemini &middot; vs antigravity &middot; pricing

---

examples

• CLAUDE.md templates · starter configs for TypeScript, Python, Rust, Next.js
• example agents · 4 agents, each teaching a different pattern
• example commands · 2 commands you can copy to any project
• handoff plugin · PreCompact context preservation
• broadcast plugin · async notifications on git events

---

how this repo works

this repo runs on its own patterns.

• 12 CI workflows · docs audit, competitive intel, community digest, freshness check, stale cleanup, dependabot, releases, plugin smoke test, PR quality gate, validation, claude responder, upstream watcher
• 11 hooks running on every session
• <$1/month CI cost · AI-powered workflows use haiku
• 0 manual maintenance · everything that doesn't require taste is automated

automation details &rarr;

---

tools i built from these patterns

these all came out of living in claude code every day. each solves a specific problem i kept hitting.

• mine · session mining to sqlite. costs, search, error memory, pattern detection
• claudemon · real-time session monitoring across projects and machines
• cc · multi-session awareness. see what other sessions are doing, send messages between them
• imessage-mcp · MCP server for read-only iMessage history. 26 tools, zero network requests

more from me

• anipotts.com/thoughts · long-form
• buttondown.com/anipotts · newsletter
• @anipottsbuilds · short-form

---

MIT &middot; built by anipotts