How real projects brief Claude Code: the commands, conventions and warnings they put in CLAUDE.md.
peterfeiCLAUDE.mdpeterfei/ai-agent-team/.claude/CLAUDE.md
位置：`.claude/agents/tech-leader.md` 2. **产品经理智能体** - `/agent product_manager` - 负责产品规划、需求分析和路线图制定 - 位置：`.claude/agents/product_manager.md` 3. **前端开发智能体** - `/agent frontend_dev` - 负责前端开发、UI实现和客户端逻辑 - 位置：`.claude/agents/frontend_dev.md` 4. **后端开发智能体** - `/agent backend_dev` - 负责后端开发、API设计和服务器端逻辑 - 位置：`.claude/agents/backend_dev.md
NgxbaCLAUDE.mdNgxba/claude-code-cli-ui/CLAUDE.md
preview # Preview production build # Type Checking bun run typecheck # Run TypeScript type checking ``` ## Architecture ### Frontend (Nuxt 3 + Vue 3) **Pages** (`app/pages/`): - `/agents` - List and manage agents - `/agents/[slug]` - Edit individual
vivekchandCLAUDE.mdvivekchand/clawmetry/CLAUDE.md
docs/MODULE_MAP.md` (generated) for every module and blueprint. TL;DR: - **Flask app** with embedded HTML/CSS/JS frontend (no build step, no npm) - **Per-feature route modules** under `routes/` — `routes/sessions.py`, `routes/usage.py`, etc. — each
alirezarezvaniCLAUDE.mdalirezarezvani/ClaudeForge/skill/CLAUDE.md
files when needed) ## Tech Stack Detection `skill/workflow.py` → `_detect_tech_stack()` reads these signals: - **Frontend** — React/Vue/Angular via `package.json`; Angular via `angular.json`; TypeScript via `tsconfig.json`. - **Backend** — Node (`package.json`), Python (`requirements.txt` / `pyproject.toml` / `setup.py
laradashboardCLAUDE.mdlaradashboard/laradashboard/CLAUDE.md
purely decorative images. ## SEO Best Practices - Every page must have a unique ` `, ` `, and ` `. - The frontend layout (`frontend-app.blade.php`) provides sensible defaults; child views override with `@section`. - Open Graph (`og:title
MicrosoftCLAUDE.mdmicrosoft/semanticworkbench/CLAUDE.md
focused context * **Library development**: Choose appropriate `PYTHON_LIBRARIES_*.md` file by functional area * **Frontend work**: Study component patterns in `WORKBENCH_FRONTEND.md` * **API development**: Follow service patterns from `WORKBENCH_SERVICE.md` * **MCP servers
jamditisCLAUDE.mdjamditis/claude-skills-journalism/CLAUDE.md
Ethical content extraction │ ├── web-ui-best-practices/ # Container queries, :has(), view transitions │ └── zero-build-frontend/ # ESM import maps, htmx, Alpine.js │ ├── # Plugin: okf-wiki (1 skill), registered in marketplace.json ├── okf-wiki
hallucinogenCLAUDE.mdhallucinogen/agent-viewer/CLAUDE.md
Express backend handling agent lifecycle, tmux integration, state detection, and SSE broadcasting - **`public/index.html`** — Entire frontend (HTML/CSS/JS) in a single file with vanilla JavaScript ### Backend Core Systems (server.js) **Agent Registry
nudgebeeCLAUDE.mdnudgebee/nudgebee/CLAUDE.md
collector-server/k8s-collector/app/` · `auto-pilot/` (+ `auto-pilot/sidecar/`) · `notifications-server/` · `llm/benchmark/` **TypeScript (1):** `app/` — frontend dashboard (Next.js) ## AI Coding Principles These principles apply to every agent (Claude, Gemini, human) working
chaitanyyaCLAUDE.mdchaitanyya/sales/CLAUDE.md
with hot reload) bun run tauri:dev # Build production app bun run tauri:build # Frontend only (Vite dev server on localhost:3000) bun run dev # Lint and format
syahiidkamilCLAUDE.mdsyahiidkamil/Software-Engineer-AI-Agent-Atlas/repos/CLAUDE.md
backend at @repos/backend (port xxxx) frontend at @repos/frontend (port yyyy
jerrywu001CLAUDE.mdjerrywu001/cc-sessions-viewer/CLAUDE.md
changes are picked up by Tauri's own dev loop. ## Architecture ### Two-side split - **Frontend** (`src/`) is a thin Vue 3 SPA. State lives in `App.vue` refs; there
aayushchCLAUDE.mdaayushch/laya/CLAUDE.md
single test cd engine && source .venv/bin/activate && pytest tests/test_cards_api.py::test_function_name -v # Frontend dev only (no Tauri shell) cd ui && npm run dev # Type check frontend cd ui && npx svelte
glebisCLAUDE.mdglebis/claude-skills/automation-advisor/CLAUDE.md
need them. --- ## Implementation Notes - Tech stack: Python Flask backend, vanilla HTML/CSS/JS frontend, Groq for voice - Current state: Design system established, recently refined for "quieter" aesthetic - Voice-enabled interface for accessibility
VasiHemanthCLAUDE.mdVasiHemanth/tokentelemetry/.claude/CLAUDE.md
github/workflows/security-audit.yml`** (CI, deterministic). Runs `npm audit --omit=dev --audit-level=high` across root / `frontend` / `website` on every PR touching a `package.json`, failing on high/critical *runtime* vulns (dev-only advisories
autonomous-aiCLAUDE.mdautonomous-ai/autonomous-os/CLAUDE.md
migration, skills, hooks, reset) | `docs/agentic/adding-agent-runtime.md` | `docs/vi/agentic/adding-agent-runtime_vi.md` | | Remote Hermes over LAN (Intern 2 as voice frontend for a Hermes server on the user's Mac, setup script, URL/token, verify) | `docs/agentic/remote-hermes.md` | `docs/vi/agentic/remote-hermes_vi.md
timwuhaotianCLAUDE.mdtimwuhaotian/the-pair/CLAUDE.md
Pair is a Tauri 2 desktop app (Rust backend, React 19 + TypeScript frontend) that orchestrates two AI agents working together on a coding task: - **Mentor** — read-only planner/reviewer - **Executor** — code
zexuanw958-svgCLAUDE.mdzexuanw958-svg/travel-plan-viz/CLAUDE.md
混合架构**：易错的机械逻辑固化为可复用 JS 引擎，视觉表现每次交给**设计步骤**重新生成。改动时别把布局/配色写死进引擎，也别把日期/导航逻辑塞给设计步骤临场生成。 - **设计步骤可插拔（无硬依赖）**：优先用 `frontend-design` 或 `huashu-design` skill（任一已安装），都没有则走 `references/design-guidelines.md` 内置准则。这样 skill 可独立分享，不强制别人先装别的 skill。别把它改回硬依赖某个外部
lovstudioCLAUDE.mdlovstudio/Ataru/CLAUDE.md
coding tool histories while retaining the legacy Lovcode CLI and storage contracts. ## Commands ```bash # Frontend development (hot reload) pnpm dev # Type check + production build pnpm build # Run Tauri desktop
milady-aiCLAUDE.mdmilady-ai/milady/CLAUDE.md
Plain language: **Eliza agents**. Exception: **Eliza Classic** plugin keeps `Eliza` (the 1966 chatbot). ## Cloud frontend visual review (REQUIRED for any UI change) > Path/mode note: `cloud-frontend` lives at `eliza/packages/cloud-frontend
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.
