How real projects brief Claude Code: the commands, conventions and warnings they put in CLAUDE.md.
mondaycomCLAUDE.mdmondaycom/vibe/CLAUDE.md
TypeScript cache (`rm -rf .rpt2_cache`) ### MCP Integration The repository includes an MCP (Model Context Protocol) server (`@vibe/mcp`) that provides intelligent assistance for working with Vibe components. Reference the MCP
ruvnetCLAUDE.mdruvnet/metaharness/kimi-k3-harness/CLAUDE.md
level implementation plan before any code. ## Commands - `doctor` — Health-check the harness: kernel load, MCP wiring, memory backend, host adapter. - `review-diff` — Review the current working diff for correctness, security
JuliusBrusseeCLAUDE.mdJuliusBrussee/cavemem/CLAUDE.md
local SQLite + vector index, and exposes them to agents through a Model Context Protocol (MCP) server and a local web viewer. The signature property of the project is that **memory
ktnytCLAUDE.mdktnyt/cclsp/CLAUDE.md
test # Full pre-publish check npm run prepublishOnly # build + test + typecheck ``` ## Architecture ### Core Components **MCP Server Layer** (`index.ts`) - Entry point that implements MCP protocol - Exposes `find_definition`, `find_references
CannerCLAUDE.mdCanner/wren-engine/.claude/CLAUDE.md
strict SQL policy mode, and optional LanceDB-backed memory layer for semantic schema/query retrieval. - **mcp-server/** — MCP server exposing Wren Engine to AI agents (Claude, Cline, Cursor). Supporting modules: `wren
henrydaumCLAUDE.mdhenrydaum/second-brain/CLAUDE.md
still lives on the store branch. There was a third, an MCP server exposing Second Brain to external MCP clients over streamable HTTP; it was never migrated — it still imported
joaoh82CLAUDE.mdjoaoh82/rustunnel/CLAUDE.md
This crate (`rustunnel/`) is the **core tunnel server + CLI client + MCP server**. Siblings in the parent meta-repo: | Sibling | Stack | Dev port | Role | |---------|-------|----------|------| | `rustunnel-web/` | Rust (Axum) + Next.js
marc-shadeCLAUDE.mdmarc-shade/world-intel-mcp/CLAUDE.md
claude.ai/code) when working with code in this repository. ## What This Is World Intelligence MCP Server: 128 tools across 30+ domains providing real-time global intelligence from free public APIs
samvallad33CLAUDE.mdsamvallad33/vestige/CLAUDE.md
operating notes, or mandatory background hooks. ## Project Shape Vestige is a local-first MCP memory server written in Rust, with a SvelteKit dashboard embedded into the release binary. The core
SixHqCLAUDE.mdSixHq/Overture/CLAUDE.md
Monorepo Structure This is an npm workspaces monorepo with two packages: - `packages/mcp-server` - The MCP server (Node.js/TypeScript) - `packages/ui` - The React frontend (React Flow + Zustand + Tailwind) ## Build & Development Commands ```bash # Install
djydeCLAUDE.mddjyde/ccmate/CLAUDE.md
Package Manager**: pnpm (required) ## Development Commands ```bash # Install dependencies pnpm install # Start development server pnpm tauri dev # Build for production pnpm build # Preview built app pnpm preview # Check TypeScript lint
tuannvmCLAUDE.mdtuannvm/codex-mcp-server/CLAUDE.md
Codex CLI ``` ### Key Files | File | Purpose | |------|---------| | `src/index.ts` | Entry point, creates and starts `CodexMcpServer` | | `src/server.ts` | MCP server setup, handles `list_tools` and `call_tool` requests | | `src/tools/definitions.ts` | MCP tool schemas (input/output
danielealbanoCLAUDE.mddanielealbano/android-remote-control-mcp/CLAUDE.md
missing. - Provide clear error messages guiding user to grant permissions. ### Logging - Log important events: MCP server start/stop, tool calls (with sanitized parameters), errors. - Use Android `Log` class with appropriate levels
emiliaprotocolCLAUDE.mdemiliaprotocol/emilia-protocol/CLAUDE.md
claiming what they show): examples/ (e.g. examples/scitt/capsule-seam-vector-v2.mjs, examples/wimse-pep/, examples/external-verification/) - Gate (productized enforcement point): packages/gate - MCP server: mcp-server/ - Capability map: docs/CAPABILITY-MAP.md - Repository disclosure boundary: docs/REPOSITORY-BOUNDARIES.md
ee0pdtCLAUDE.mdee0pdt/Godot-MCP/CLAUDE.md
CLAUDE.md - Godot MCP Project Guidelines ## Build & Run Commands - **Server Build**: `cd server && npm run build` - **Server Start**: `cd server && npm run start` - **Server Dev Mode**: `cd server
dbt-labsCLAUDE.mddbt-labs/dbt-mcp/CLAUDE.md
tool_names) - Tool categories: `discovery/`, `semantic_layer/`, `dbt_cli/`, `dbt_codegen/`, `dbt_admin/`, `lsp/`, `mcp_server_metadata/` - Prompts (tool descriptions): `src/dbt_mcp/prompts/` - Config: `src/dbt_mcp/config/` - Tests: `tests/unit/`, `tests/integration/` ## Tool Architecture Tools follow
alpbahadurCLAUDE.mdalpbahadur/49-IDE/CLAUDE.md
HTML files reference .min.js versions so this must be done before restarting the cloud server. REBUILDING AGENT TARBALL: After changing any agent code (agent/src/, agent/services/, agent/bin/), rebuild the tarball that
matiasbattocchiaCLAUDE.mdmatiasbattocchia/open-bsp-api/CLAUDE.md
filter by timestamp. ### Querying HTTP-level logs (status codes, execution time) Use the Supabase MCP server `get_logs` tool with `service: "edge-function"`. This returns invocation metadata (status code, execution
StuMasonCLAUDE.mdStuMason/coolify-mcp/CLAUDE.md
definition 4. **src/**tests**/mcp-server.test.ts** - Add mocked test ### Key Files - **src/index.ts** - Entry point, starts MCP server - **src/lib/coolify-client.ts** - HTTP client wrapping Coolify REST API - **src/lib/mcp-server.ts** - MCP tool definitions and handlers, plus
xianyu110CLAUDE.mdxianyu110/awesome-claudcode-tutorial/CLAUDE.md
三种模式详解 ├── 05-terminal-commands.md # 终端命令执行 ├── 06-plan-mode.md # 规划模式 ├── 07-permissions.md # 权限管理 ├── 08-background-tasks.md # 后台任务 ├── 09-rewind.md # 版本回滚 ├── 10-image-processing.md # 图片处理 ├── 11-mcp-setup.md # MCP Server 安装 ├── 12-resume.md # 恢复历史会话 ├── 13-mcp-design.md # MCP 还原设计稿 ├── 14-context-management.md # 上下文管理 ├── 15-claude-md.md # CLAUDE.md 项目记忆 ├── 16-hooks.md # Hook 系统
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.
