# revibe-mcp MCP server for [Revibe](https://revibe.codes) codebase analysis. Works with Claude Desktop, Cursor, Windsurf, and any MCP-compatible client. Analyze any GitHub repo — get architecture, file roles, execution flows, system design Q&A, and structured agent context. ## Quick Start (Node.js — zero install) Add to your MCP client config: ```json { "mcpServers": { "revibe": { "command": "npx", "args": ["revibe-mcp"] } } } ``` Then ask your AI assistant to "log in to Revibe" — it will open your browser for a one-time signup. ## Quick Start (Python) ```bash pip install revibe-mcp revibe-mcp-auth login ``` ```json { "mcpServers": { "revibe": { "command": "revibe-mcp" } } } ``` ## Where to add the config | Client | Config location | |--------|----------------| | **Claude Desktop** | `~/Library/Application Support/Claude/claude_desktop_config.json` | | **Claude Code** | `.mcp.json` in your project root | | **Cursor** | Settings > MCP Servers > Add | | **Windsurf** | MCP config in settings | ## Auth Two options: **Option A: Browser login (recommended)** ```bash # Node.js npx revibe-mcp-auth login # Python revibe-mcp-auth login ``` Opens your browser — sign up or log in, click "Authorize". API key saved automatically to `~/.config/revibe/credentials.json`. **Option B: Manual API key** Get a key from [app.revibe.codes/settings](https://app.revibe.codes/settings), then pass it via env: ```json { "mcpServers": { "revibe": { "command": "npx", "args": ["revibe-mcp"], "env": { "REVIBE_API_KEY": "rk_live_your_key_here" } } } } ``` ## Tools | Tool | Description | |------|-------------| | `revibe_login` | Log in via browser — one-time setup | | `analyze_repo` | Submit a GitHub repo for analysis (auto-detects from git remote) | | `check_status` | Check progress of an analysis job | | `get_summary` | Architecture, modules, patterns overview | | `get_section` | Drill into architecture, file_roles, system_design_qa, etc. | | `get_agent_context` | Full structured JSON optimized for AI agents | ## Usage Once configured, just ask your AI assistant: - "Analyze this repo with Revibe" - "What's the architecture of github.com/user/repo?" - "Get the agent context for this codebase" The tools handle auto-detection, polling, and structured output automatically. ## CLI Commands ```bash # Node.js npx revibe-mcp-auth login # Log in via browser npx revibe-mcp-auth status # Show current auth status npx revibe-mcp-auth logout # Remove saved credentials # Python revibe-mcp-auth login revibe-mcp-auth status revibe-mcp-auth logout ``` ## Claude Code Skill If you use [Claude Code](https://claude.ai/code), you can install Revibe as a slash command skill instead of (or in addition to) the MCP server: ```bash # Copy the skill file mkdir -p ~/.claude/skills/revibe curl -o ~/.claude/skills/revibe/SKILL.md \ https://raw.githubusercontent.com/selvatuple/revibe-mcp/main/skills/claude-code/SKILL.md ``` Then use `/revibe` or `/revibe github.com/user/repo` inside Claude Code. The skill file is also available at [`skills/claude-code/SKILL.md`](skills/claude-code/SKILL.md). ## Packages & Distribution | Method | Install | |--------|---------| | **npm** (zero install) | `npx revibe-mcp` | | **PyPI** | `pip install revibe-mcp` | | **MCP Registry** | [registry.modelcontextprotocol.io](https://registry.modelcontextprotocol.io) — search "revibe" | | **Claude Code Skill** | Copy [`skills/claude-code/SKILL.md`](skills/claude-code/SKILL.md) to `~/.claude/skills/revibe/` | ## License MIT