Give Claude Code and Cursor a shared, auto-updating map of your codebase

说明与用途

搭建共享 MCP 服务器，自动更新代码库知识图谱，让 AI 助手无需逐文件读取即可查询代码关系。

问题

AI coding assistants struggle to locate code elements across large repositories, leading to token waste and incomplete context.

结果

The shared MCP server keeps the code graph in sync with the main branch, eliminates per‑user setup, and allows assistants to query the graph for relationships, paths, and impact analysis without re‑reading files.

工作流

• Clone the repository using a read‑only token
• Run graphify extract --code-only to generate graph.json
• Serve the graph with graphify.serve --transport http on /mcp
• Configure hourly refresh to pull, rebuild, and atomically swap graph.json
• Expose the server behind HTTPS and secure with an API key
• Add the MCP server to Claude Code or Cursor with the API key header

证据

Graphify fixes that by turning a codebase into a knowledge graph: functions, classes, modules and the edges between them, parsed locally with tree‑sitter.

来源

Graphify already ships an HTTP transport for its MCP server, so the missing piece is just the plumbing: clone, build, serve, refresh, secure.

来源

The server is reachable on the internet, and the API key is what keeps your code map private.

来源