How real projects brief Claude Code: the commands, conventions and warnings they put in CLAUDE.md.
IvanMurzakCLAUDE.mdIvanMurzak/Unity-MCP/Unity-MCP-Plugin/CLAUDE.md
clients. ## Build / run - **Open**: `Unity-MCP-Plugin` folder in Unity Editor (compiles automatically) - **Tests**: Unity Test Runner (`Window > General > Test Runner`) — EditMode in `Packages/com.ivanmurzak.unity.mcp/Tests/Editor`, PlayMode in `Packages/com.ivanmurzak.unity.mcp/Tests/Runtime` - **MCP Inspector**: `Commands/start_mcp_inspector.bat
VercelCLAUDE.mdvercel-labs/just-bash/CLAUDE.md
check pnpm lint:fix # Fix lint errors (biome) pnpm knip # Check for unused exports/dependencies # Testing pnpm test:run # Run ALL tests (including spec tests) pnpm test:unit # Run unit tests
archestra-aiCLAUDE.mdarchestra-ai/archestra/ai-labs/CLAUDE.md
skill/tool that solves it -- finding the right tool is itself a key capability under test, so naming it hands the model the answer. A prompt names only the **delivery-protocol
httprunnerCLAUDE.mdhttprunner/httprunner/CLAUDE.md
when working with code in this repository. ## Project Overview HttpRunner v5 is a comprehensive testing framework written in Go that supports API testing, load testing, and UI automation across multiple
txn2CLAUDE.mdtxn2/kubefwd/CLAUDE.md
/cmd/kubefwd/kubefwd.go # Build with version information go build -ldflags "-X main.Version=dev" -o kubefwd ./cmd/kubefwd/kubefwd.go ``` ### Testing ```bash # Run all tests go test ./... # Run tests for a specific package go test ./pkg/fwdport
melgarafaelCLAUDE.mdmelgarafael/DeskcommCRM/CLAUDE.md
coisa.** - [`docs/harness-audit.md`](docs/harness-audit.md) — onde a verificação tem buraco. Importante: `pnpm gov:verify` **não** cobre `test:db` nem `test:e2e` — verde ali não é prova para mudança de schema
nyldnCLAUDE.mdnyldn/claude-octopus/CLAUDE.md
skills, agents, hooks) - User-facing documentation (README.md, CHANGELOG.md, docs/) - Build/config files (package.json, tsconfig.json, .gitignore) - Test files in `tests/` directory ### Enforcement If you need to track progress or create working files
bethingtonCLAUDE.mdbethington/ghidra-mcp/CLAUDE.md
near zero with AI. Do the whole thing. Do it right. Do it with tests. Do it with documentation. Do it so well that Ben is genuinely impressed — not politely
StripeCLAUDE.mdstripe/stripe-php/.claude/CLAUDE.md
stripe-php ## Testing - Run all tests: `just test` - Run specific test: `just test --filter testMethodName` - Run specific test file: `just test tests/Stripe/SomeTest.php` - Tests use PHPUnit ## Formatting & Linting - Format: `just format
BenedictKingCLAUDE.mdBenedictKing/ccx/backend-go/CLAUDE.md
协议转换、多渠道调度、会话管理、指标与日志记录、配置热重载。 ## 启动命令 ```bash make dev make run make build make test make test-cover make fmt make lint make deps ``` 说明：实际命令以 `backend-go/Makefile` 为准
BenedictKingCLAUDE.mdBenedictKing/ccx/CLAUDE.md
后端开发 (backend-go/) make dev # 热重载开发模式 make run # 复制前端产物后直接运行 make build # 构建生产版本 make test # 运行所有测试 make test-cover # 测试 + 覆盖率报告 make fmt # 格式化代码 make lint # 代码检查 make deps # 更新依赖 # 前端开发
badrisnarayananCLAUDE.mdbadrisnarayanan/antigravity-claude-proxy/CLAUDE.md
browser # headless/manual code input npm run accounts:list npm run accounts:verify npm test # requires server running on port 8080 node tests/run-all.cjs # run matching tests only node tests/test-strategies.cjs # strategy unit
matt1398CLAUDE.mdmatt1398/claude-devtools/CLAUDE.md
Type checking - `pnpm lint:fix` - Lint and auto-fix - `pnpm format` - Format code - `pnpm test` - Run all vitest tests - `pnpm test:watch` - Watch mode - `pnpm test:coverage` - Coverage report - `pnpm
AnthropicCLAUDE.mdanthropics/anthropic-sdk-python/CLAUDE.md
default branch. Those branches get force-pushed, so rebase with `git rebase --onto `. ## Build, test, lint The entry points are `./scripts/{bootstrap,format,lint,test}`. Always use them to format
nowork-studioCLAUDE.mdnowork-studio/notfair-plugin/CLAUDE.md
workflow recurs or the bug pattern is reusable, encode it as a test, guardrail, skill, or automation. ## Agent entry points (single sources of truth) - **`AGENTS.md`** — the universal skill resolver. Every
nowork-studioCLAUDE.mdnowork-studio/notfair-plugin/notfair/CLAUDE.md
migrations or legacy/compat code. Verification is: 1. `pnpm typecheck` — must be clean. 2. `pnpm test` — vitest; must be clean. Tests live next to the code they cover (`src/lib/foo.ts` → `src/lib/foo.test.ts`). Default
dromaraCLAUDE.mddromara/liteflow/CLAUDE.md
脚本 / 特性维度划分，见下文「测试基础设施」）。 > **注意：** 根 pom 的 surefire 默认 `skipTests=true`，直接 `mvn test` 不会真正执行测试，需要显式加 `-DskipTests=false`。 ```bash # 运行所有测试 mvn test -DskipTests=false # 运行指定模块的测试（liteflow-testcase-el/ 下有
gadievronCLAUDE.mdgadievron/raptor/CLAUDE.md
with `--help`. --- ## COMMANDS /project - Project management — `libexec/raptor-project-manager [args]` /scan /fuzz /web /codeql /analyze - Security testing — `python3 raptor.py ` /agentic - Scan → dedup → analysis pipeline — `libexec/raptor-agentic --repo ` /exploit /patch - Generate PoCs and fixes
glitternetworkCLAUDE.mdglitternetwork/pinme/CLAUDE.md
build # Production build (esbuild → dist/index.js) npm run dev # Dev build npm run test # Unit/integration tests (Vitest) npm run test:cli # Real CLI black-box tests against dist/index.js npm run verify
xintaofeiCLAUDE.mdxintaofei/codeg/CLAUDE.md
intl - **数据库**: SeaORM + SQLite - **包管理器**: pnpm ## 代码检查与测试（任务完成后进行必要的检查） ### 前端 ```bash pnpm lint . # lint pnpm test # vitest 全跑（CI 用同一条命令） pnpm test:watch # 开发时增量重跑 pnpm test:coverage # 覆盖率报告（输出到 coverage/index.html） pnpm
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.
