What problem does repo-graph solve?
It replaces the inefficient “grep, read, grep, read” loop by giving the AI a structural map of the codebase. In one benchmark, it reduced tokens from 75K to 30K and time from 4m36s to ~30s.
@James-Chahwan
Structural graph memory for AI coding assistants. repo-graph maps your codebase entities, relationships, and feature flows so the model navigates to the right files instead of reading everything first. Tree-sitter, 20+ languages, frontend to backend and more, any MCP client.
使用下面的配置,将此服务器添加到你的 MCP 客户端。
{
"mcpServers": {
"repo-graph": {
"command": "uvx",
"args": [
"mcp-repo-graph",
"--repo",
"."
]
}
}
}工具列表暂未提供
当前收录暂时没有可展示的工具列表,不代表服务没有工具。请查阅项目文档了解支持的能力。
点击上方按钮,在临时沙箱中运行此服务器并实时获取工具列表。
Repo Graph is an MCP server that builds a structural graph of a codebase — entities, their connections, and feature flows from frontend to backend — for AI coding assistants. It uses Rust and tree-sitter to handle 20+ languages and frameworks cross-stack, allowing the AI to query the graph for relevant files instead of reading the entire repository.
Install and run with uvx mcp-repo-graph --repo . (zero-install via uvx). Add it to any MCP client (Claude Code, Cursor, Windsurf, Codex, Gemini CLI, etc.) by configuring the MCP server with the same command.
It replaces the inefficient “grep, read, grep, read” loop by giving the AI a structural map of the codebase. In one benchmark, it reduced tokens from 75K to 30K and time from 4m36s to ~30s.
It handles 20+ languages and frameworks, including cross-stack support (e.g., Go + Angular). The exact list isn’t given, but it uses tree-sitter for broad coverage.
Installation is zero-install with `uvx mcp-repo-graph --repo .`. For Claude Code, use `claude mcp add repo-graph -- uvx mcp-repo-graph --repo .`. For other MCP clients, add the same command to their MCP server configuration.
It works with any MCP client: Cursor, Windsurf, Codex, Gemini CLI, Antigravity, Claude Desktop, and Claude Code.
It offers 11 tools: generate, status, dense_text, flow, trace, impact, neighbours, activate, find, graph_view, and reload.
Put an end to code hallucinations! GitMCP is a free, open-source, remote MCP server for any GitHub project
MCP server that enables AI assistants to interact with Linear project management system through natural language, allowing users to retrieve, create, and update issues, projects, and teams.
🚀 All-in-one MCP server with AI search, RAG, and multi-service integrations (GitLab/Jira/Confluence/YouTube) for AI-enhanced development workflows. Folk from https://github.com/nguyenvanduocit/all-in-one-model-context-protocol
Learn how to use MCP Servers with GitHub Copilot
MCP servers for the Atlassian products (Bitbucket, Confluence, JIRA) of the Data Center version
评论