MPS
JetBrains/MPS/CLAUDE.md
edits corrupt models. Use the MPS MCP tools (`mps_mcp_*`) instead. If the MPS MCP server is not connected, ask the user to start MPS and enable it; a connected
How real projects brief Claude Code: the commands, conventions and warnings they put in CLAUDE.md.
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
JetBrains/MPS/CLAUDE.md
edits corrupt models. Use the MPS MCP tools (`mps_mcp_*`) instead. If the MPS MCP server is not connected, ask the user to start MPS and enable it; a connected
manzaltu/claude-code-ide.el/CLAUDE.md
claude-code-ide-mcp-handlers.el` - MCP tool implementations (file ops, ediff, diagnostics) **Support Files:** - `claude-code-ide-mcp-server.el` - HTTP-based MCP tools server framework - `claude-code-ide-mcp-http-server.el` - HTTP transport implementation - `claude-code-ide-emacs-tools.el` - Emacs tools: xref, project info, imenu - `claude-code-ide-diagnostics.el` - Flycheck
shrimbly/node-banana/CLAUDE.md
with each message → `/api/agent/chat` runs the turn on the chosen harness → tools edit a server-side draft and emit resolved graph ops → the browser applies them with `applyAgentGraphOps` (one undo
LewisWJackson/tradingview-mcp-jackson/CLAUDE.md
study by default (pass `max_labels` to override) ## Architecture ``` Claude Code ←→ MCP Server (stdio) ←→ CDP (localhost:9222) ←→ TradingView Desktop (Electron) ``` Pine graphics path: `study._graphics._primitivesCollection.dwglines.get('lines').get(false)._primitivesDataById
ghostwright/ghost-os/CLAUDE.md
Build & Run ```bash swift build # Requires Swift 6.2+ (swiftly), macOS 14+ .build/debug/ghost mcp # Start MCP server (stdio) .build/debug/ghost setup # Interactive first-run wizard .build/debug/ghost doctor # Diagnostic health check .build/debug/ghost status
Simpleyyt/ai-manus/CLAUDE.md
drive the sandbox's headless Chrome (viewable via VNC→websockify→NoVNC); `mcp.py` loads external MCP servers from a mounted `mcp.json` (see `mcp.json.example`). Plan progress for the UI comes from Flow
perminder-klair/subwave/web/CLAUDE.md
server read. `setQueryData` stamps `dataUpdatedAt` too, so a decision that needs what the server said takes it from the fetch's own success (`dash/never-play.ts`, #2018). - Whole-document editors (`SchedulePanel`, `WebhooksPanel
svnscha/mcp-windbg/CLAUDE.md
main(): CLI argument parsing, picks the transport __main__.py module entry point server.py MCP server: session lifecycle + call_tool dispatch server_api.py Tool param models + list_tools + packaged prompt handlers cdb_session.py
No other project uses it yet
leonardsellem/n8n-mcp-server/CLAUDE.md
This will be populated once the project structure is established. Key areas to document: - MCP server implementation details - n8n integration patterns - Configuration management - API endpoints and protocols
archestra-ai/OpenAPPA/appa-runtime/CLAUDE.md
runtime API and internal `Session` event model, the `hooks` dispatcher, the HTTP server, the MCP endpoint, the externals, the builtin modules (`builtins.rs` — stock implementations plus the `--modules-dir` loader over
actionbook/actionbook/CLAUDE.md
Module Structure ``` actionbook/ ├── packages/ │ ├── js-sdk/ # @actionbookdev/sdk - JavaScript SDK with tool definitions │ ├── mcp/ # @actionbookdev/mcp - MCP Server (standalone, publishable to npm) │ ├── cli/ # @actionbookdev/cli - Command line interface (Rust CLI + npm wrapper) │ └── tools
docker/mcp-gateway/CLAUDE.md
Gateway** - a CLI plugin that enables easy and secure running of Model Context Protocol (MCP) servers through Docker containers. The plugin acts as a gateway between AI clients and containerized
No other project uses it yet
amd/gaia/CLAUDE.md
agent.py` | Base `Agent` class | Inherit for all new agents | | `mcp_agent.py` | `MCPAgent` mixin | Add MCP protocol support | | `api_agent.py` | `ApiAgent` mixin | Add OpenAI-compatible API exposure | | `tools.py` | `@tool` decorator, registry | Register
Steady5 mentions in 2 weeks
No other project uses it yet
Flux159/mcp-server-kubernetes/CLAUDE.md
Desktop testing, point to local `dist/index.js` build ## Architecture Overview This is an MCP (Model Context Protocol) server that provides Kubernetes cluster management capabilities. The server connects to Kubernetes clusters
silverstein/minutes/CLAUDE.md
appropriate app bundle - TCC-sensitive desktop work (hotkeys, Screen Recording, Input Monitoring, Accessibility): `./scripts/install-dev-app.sh` - MCP server changes: `cd crates/mcp && npm run build` (compiles TS server + builds UI, then restart MCP
No other project uses it yet
hyperbrowserai/HyperAgent/CLAUDE.md
ranking → action execution | | `agent/messages/` | LLM prompt construction (system prompt, input/output format, action examples) | | `agent/mcp/client.ts` | MCP server integration — registers remote tools as agent actions | | `cdp/` | Chrome DevTools Protocol: frame graph tracking
NanoNets/docstrange/CLAUDE.md
app.nanonets.com/#/keys - Pass via `api_key` parameter or `NANONETS_API_KEY` env var ## MCP Server Integration The repository includes an MCP server for Claude Desktop integration (local development only): ### Setup
six2dez/burp-ai-agent/CLAUDE.md
/2/3. - **Target**: Burp Suite Community + Professional 2023.12+, cross-platform (macOS / Linux / Windows). - **Network**: MCP server binds to `127.0.0.1` by default; external access requires explicit opt-in + bearer token + optional
akfamily/aktools/CLAUDE.md
person auth belongs in a gateway such as OAuth2 Proxy. - `aktools/mcp_server.py` — MCP server, a required dependency pinned `mcp>=2.3,<3` (the v2 SDK, `MCPServer`, not v1's `FastMCP`; majors
Open-Source-Legal/OpenContracts/CLAUDE.md
commit hooks) pre-commit run --all-files ``` ### Frontend (React/TypeScript) ```bash cd frontend # Start development server (proxies to Django on :8000) yarn start # Run unit tests (Vitest) - watches by default yarn
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.