How real projects brief Claude Code: the commands, conventions and warnings they put in CLAUDE.md.
lobehubCLAUDE.mdlobehub/lobehub/e2e/CLAUDE.md
目录运行 cd e2e # 运行特定标签的测试 BASE_URL=http://localhost:3006 \ DATABASE_URL=postgresql://postgres:postgres@localhost:5433/postgres \ pnpm exec cucumber-js --config cucumber.config.js --tags "@AGENT-CHAT-001" # 调试模式（显示浏览器） HEADLESS=false
microCLAUDE.mdmicro/go-micro/CLAUDE.md
Client-side load balancing ├── server/ # RPC server ├── service/ # Service interface + profiles ├── store/ # Data persistence (Postgres, NATS KV) ├── transport/ # Network transport ├── wrapper/ # Middleware (auth, trace, metrics) ├── examples/ # Working examples └── internal
CannerCLAUDE.mdCanner/WrenAI/.claude/CLAUDE.md
with all optional extras (incl. memory) just install-extra # e.g. just install-extra postgres just install-memory # memory extra (lancedb + sentence-transformers) just install-local # engine dev: uv sync + build
open-metadataCLAUDE.mdopen-metadata/OpenMetadata/CLAUDE.md
generated/**` — regen, never hand-edit generated | | `migrations.md` | `bootstrap/sql/**` — append-only, native path, MySQL+Postgres, idempotent | ### Repo coding conventions (read before writing non-trivial code) - `docs/design-patterns.md` — the design patterns this codebase
A file Claude Code reads at the start of every session. It holds the commands, conventions and warnings the agent needs for this project.
At the repository root. Claude Code also reads CLAUDE.md files in subdirectories when it works there.
Build and test commands, the project's layout, conventions that aren't obvious from the code, and mistakes to avoid. Short files tend to work better than long ones.
Claude Code reads CLAUDE.md; most other agents read AGENTS.md. Many projects keep one and point the other at it.
