contextflow-mcp

MCP server for Contextflow — semantic code search and AI-powered Q&A over vectorized repositories, directly inside Claude.

🚀 What is this?

contextflow-mcp is a Model Context Protocol server that connects Claude to a Contextflow snapshot — a vectorized index of your codebase. Once connected, Claude can:

• 🔍 Semantically search your code
• 💬 Answer natural language questions grounded in actual source code
• 📦 Inspect snapshot metadata (files indexed, status, branch)

---

🛠️ Tools exposed

Tool	Description
get_context_info	Returns metadata about the snapshot: name, branch, status, files and chunks indexed
search_context	Semantic search — returns the most relevant code chunks with file paths and line numbers
ask_context	AI Q&A — asks a question and gets an answer grounded in real source code

---

⚙️ Setup

1. Install dependencies

npm install

2. Build

npm run build

3. Configure Claude

Add the following to your Claude MCP config (claude_desktop_config.json or equivalent):

{
  "mcpServers": {
    "contextflow": {
      "command": "node",
      "args": ["/absolute/path/to/dist/index.js"],
      "env": {
        "CONTEXTFLOW_TOKEN": "YOUR_SNAPSHOT_PUBLIC_TOKEN",
        "CONTEXTFLOW_API_URL": "http://localhost:3001"
      }
    }
  }
}

See claude-config-example.json for a ready-to-copy template.

---

🌍 Environment Variables

Variable	Required	Default	Description
CONTEXTFLOW_TOKEN	✅ Yes	—	Public token of the Contextflow snapshot
CONTEXTFLOW_API_URL	No	http://localhost:3001	Base URL of the Contextflow API

---

📁 Project Structure

contextflow-mcp/
├── src/
│   └── index.ts          # MCP server source (TypeScript)
├── dist/
│   └── index.js          # Compiled output (Node.js)
├── package.json
├── tsconfig.json
└── claude-config-example.json

---

🧑‍💻 Development

npm run dev    # watch mode (tsc --watch)
npm run build  # single compile
npm start      # run compiled server

---

📄 License

MIT