GitHub Stars MCP Server
Enables natural language search and querying of your GitHub starred repositories.
README
GitHub Stars MCP Server
A Cloudflare-powered MCP (Model Context Protocol) Server that allows you to search and query your GitHub starred repositories using natural language.
English | 简体中文
Overview
This project creates a searchable database of your GitHub starred repositories by:
- Fetching all your starred repositories using the GitHub API
- Extracting and processing the README files from each repository
- Uploading the processed data to Cloudflare R2 storage
- Using Cloudflare AutoRAG to create embeddings for efficient searching
- Exposing a MCP Server that allows querying these repositories via natural language
Features
- Automatically fetches and processes GitHub starred repositories
- Scheduled weekly updates via GitHub Actions
- Stores repository metadata and README content
- Provides semantic search capabilities through Cloudflare AutoRAG
- Exposes a MCP-compatible API for integration with AI agents
Setup
Prerequisites
- Node.js (v22 recommended)
- PNPM package manager
- GitHub Personal Access Token with
reposcope - Cloudflare account
Configuration
-
Clone this repository
-
Set up Cloudflare R2:
- Create a R2 bucket
- Configure R2 access credentials
-
Configure GitHub Secrets for the CI/CD workflow:
GH_TOKEN: GitHub token for fetching starred repositoriesR2_ACCOUNT_ID: Cloudflare account IDR2_ACCESS_KEY_ID: R2 access keyR2_SECRET_ACCESS_KEY: R2 secret keyR2_BUCKET: R2 bucket name
-
Configure Cloudflare AutoRAG:
- Create an AutoRAG instance in Cloudflare
- Set the
AUTO_RAG_NAMEenvironment variable in your Cloudflare Worker MCP_API_KEY: API key for securing MCP server access (generate a secure random string)
Local Development
To develop locally:
# Install dependencies
pnpm install
# Fetch your GitHub stars locally
pnpm dev:stars
# Run MCP server locally
pnpm dev:mcp
Deployment
Deploy to Cloudflare Workers:
pnpm deploy
The GitHub Action will automatically:
- Run weekly to update your starred repositories
- Upload the processed files to R2
- Rebuild the AutoRAG index
Usage
Once deployed, you can interact with the MCP Server using any MCP-compatible client:
API Key Authentication
The MCP server requires API key authentication for security. Include your MCP_API_KEY in the request headers:
Authorization: Bearer your-mcp-api-key
Endpoint
Streamable HTTP: https://your-worker-url.workers.dev
API Reference
MCP Tool: search_github_stars
Searches through your starred GitHub repositories.
Parameters:
query(string): Natural language query to search repositories
Response:
- JSON result containing matching repositories and relevant README content
推荐服务器
Baidu Map
百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。
Playwright MCP Server
一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。
Magic Component Platform (MCP)
一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。
Audiense Insights MCP Server
通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。
VeyraX
一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。
graphlit-mcp-server
模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。
Kagi MCP Server
一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
Exa MCP Server
模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。