🐙 GitHub Analyzer Pro MCP服务器v2.0
A. 生产准备就绪 模型上下文协议（MCP）服务器 28+强大的工具 用于使用FastMCP构建的全面GitHub分析。
](https://www.python.org/downloads/)  ](https://docs.github.com/rest) 
______________________________________________________________________
✨ 完整功能集
📦 存储库分析 （8工具）
- get_pository_info -包含统计信息的完整仓库元数据
- get_pository_语言 -按百分比分列的语言
- get_pository_conibutors -顶级贡献者排名
- get_pository_tags -发布标签列表
- get_pository_brances -具有保护状态的分支机构信息
- get_pository_stats -全面分析
- get_pository_traft -视图和克隆（需要权限）
- get_pository_社区 -健康评分和社区文件
📝 承诺与历史 （3个工具）
- get_pository_commits -与作者分享历史
- get_commit_details -深入了解具体的提交
- comparecommits -比较分支或提交
🐛 问题和拉取请求 （3个工具）
- 获取库存问题 -按州、标签筛选
- get_pull_requests -具有合并状态的PR详细信息
- 获取问题注释 -评论帖子
📁 文件和内容 （4工具）
- 获取_文件_内容 -从repo读取任何文件
- get_directory_contents -浏览目录结构
- get_pository_readme -快速访问README
- 搜索代码 -跨存储库查找代码
🔍 搜索与发现 （5工具）
- search_repositories -高级回购搜索
- 搜索用户 -查找用户和组织
- 搜索主题 -发现GitHub主题
- get_trending_repositories -按语言分类的热门存储库
- 获取_渲染_开发人员 -活跃的开发人员
👤 用户和组织 （4工具）
- get_user_profile -完整的用户信息
- get_user_repositories -用户的公共仓库
- 获取用户活动 -最近的活动提要
- 获取组织信息 -组织详细信息
📦 发布与套餐 （2个工具）
- 获取租赁 -所有带有下载统计信息的版本
- get_latest_release -最新发布详情
⚙️ 工作流程和操作 （2个工具）
- get_workflows -列出GitHub操作工作流
- get_workflow_run -最近的工作流执行历史记录
🛠️ 公用事业 （2个工具）
- get_rate_limit -检查API费率限制状态
- 服务器信息 -服务器功能和统计数据
______________________________________________________________________
🚀 快速开始
1.安装依赖项
pip install fastmcp httpx python-dotenv2.设置GitHub Token（推荐）
无令牌： 60个请求/小时\ 使用令牌： 5000次请求/小时+完全搜索权限
# Linux/Mac
export GITHUB_TOKEN='ghp_your_token_here'
# Windows PowerShell
$env:GITHUB_TOKEN='ghp_your_token_here'
# Or create .env file
echo "GITHUB_TOKEN=ghp_your_token_here" > .env获取您的令牌： https://github.com/settings/tokens
- 所需范围：
public_repo,read:user - 可选：
repo（用于私有存储库）
3.运行服务器
python github_server.py预期产量：
============================================================
🐙 GitHub Analyzer Pro MCP Server v2.0.0
============================================================
📡 Server Configuration:
Transport: HTTP
Host: 0.0.0.0
Port: 8001
Timeout: 30.0s
🔑 Authentication:
GitHub Token: ✓ Configured
Rate Limit: 5000/hr
📊 Available Tools: 28
• Repository Analysis (8 tools)
• Commits & History (3 tools)
• Issues & PRs (3 tools)
• Files & Content (4 tools)
• Search & Discovery (5 tools)
• Users & Organizations (4 tools)
• Releases & Packages (2 tools)
• Workflows & Actions (2 tools)
• Utilities (2 tools)
✨ Production Features:
✓ Enhanced error handling
✓ Request caching
✓ Rate limit monitoring
✓ Community health metrics
✓ Traffic analytics
✓ GitHub Actions support
🚀 Server Status: READY
Access at: http://localhost:8001
============================================================______________________________________________________________________
📚 使用示例
🔍 高级存储库分析
# Get comprehensive stats
{
"owner": "facebook",
"repo": "react"
}
# Response includes:
# - Stars, forks, watchers
# - Language breakdown
# - Commit activity
# - Health score
# - Community files
# - Traffic data (if authorized)📊 语言的分布
# Get language percentages
{
"owner": "microsoft",
"repo": "vscode"
}
# Returns:
# - TypeScript: 45.2%
# - JavaScript: 30.1%
# - CSS: 15.7%
# - etc.👥 主要贡献者
# See who contributes most
{
"owner": "torvalds",
"repo": "linux",
"limit": 10
}🔄 比较分支机构
# See what changed between branches
{
"owner": "python",
"repo": "cpython",
"base": "3.11",
"head": "main"
}
# Shows:
# - Commits ahead/behind
# - Files changed
# - Line additions/deletions🐛 按标签筛选问题
# Find specific issues
{
"owner": "microsoft",
"repo": "vscode",
"state": "open",
"labels": "bug,high-priority"
}📦 发布管理
# Get latest release
{
"owner": "nodejs",
"repo": "node"
}
# Returns:
# - Version tag
# - Release notes
# - Download links
# - Asset download counts⚙️ 监控GitHub操作
# Check CI/CD status
{
"owner": "rust-lang",
"repo": "rust"
}
# See:
# - All workflows
# - Recent runs
# - Success/failure status👤 用户活动源
# What's a user been up to?
{
"username": "gvanrossum",
"limit": 20
}
# Shows:
# - Recent commits
# - PRs created
# - Issues opened
# - Repos starred🔥 发现趋势
# Hot Python repos this week
{
"language": "python",
"since": "weekly"
}📈 社区健康
# Check repo health
{
"owner": "django",
"repo": "django"
}
# Returns:
# - Health percentage
# - README quality
# - Has CODE_OF_CONDUCT
# - Has CONTRIBUTING guide
# - Issue templates______________________________________________________________________
🔧 连接到克劳德桌面
配置
macOS：\ ~/Library/Application Support/Claude/claude_desktop_config.json
窗户：\ %APPDATA%\Claude\claude_desktop_config.json
Linux：\ ~/.config/Claude/claude_desktop_config.json
{
"mcpServers": {
"github-pro": {
"url": "http://localhost:8001/mcp/v1"
}
}
}远程服务器：
{
"mcpServers": {
"github-pro": {
"url": "http://your-server-ip:8001/mcp/v1"
}
}
}重新启动克劳德桌面 配置后！
______________________________________________________________________
💬 与克劳德交谈
一旦连接，问克劳德这样的问题：
"Analyze the React repository - give me complete statistics"
"Show me the language breakdown for microsoft/vscode"
"Who are the top 10 contributors to the Linux kernel?"
"Compare the main and develop branches of my-org/my-repo"
"Find all open high-priority bugs in facebook/react"
"What's the latest release of Node.js? Show download counts"
"Are GitHub Actions passing for rust-lang/rust?"
"What has Guido van Rossum been working on lately?"
"Show me trending Rust repositories this month"
"Check the community health score for django/django"
"Search for async functions in the FastAPI codebase"
"Get traffic stats for my repository" (requires auth)
"Show me all workflows and their status for my-org/api"______________________________________________________________________
🏭 生产特点
✅ 增强的错误处理
- 详细的错误消息
- 速率限制检测
- 身份验证失败处理
- 网络超时管理
- 找不到资源处理
✅ 性能优化
- 请求超时控制（30s）
- 响应缓存（5分钟TTL）
- 高效的批处理请求
- 最大结果限制
✅ 安全
- 基于令牌的身份验证
- 安全环境变量
- 无令牌日志记录
- 符合费率限制
✅ 监控与调试
- 速率限制跟踪
- 请求/响应日志记录
- 服务器运行状况
- 详细的工具统计
✅ 数据质量
- 响应验证
- 安全的JSON编码
- 截断大量响应
- 二进制文件处理
______________________________________________________________________
🐳 Docker部署
Dockerfile
FROM python:3.10-slim
WORKDIR /app
# Install dependencies
COPY requirements.txt .
RUN pip install --no-cache-dir -r requirements.txt
# Copy server
COPY github_server.py .
# Environment
ENV GITHUB_TOKEN=""
ENV PYTHONUNBUFFERED=1
# Health check
HEALTHCHECK --interval=30s --timeout=10s --start-period=5s --retries=3 \
CMD python -c "import httpx; httpx.get('http://localhost:8001/mcp/v1/health')"
EXPOSE 8001
CMD ["python", "github_server.py"]需求.txt
fastmcp>=2.14.0
httpx>=0.24.0
python-dotenv>=1.0.0构建与运行
# Build image
docker build -t github-mcp-pro .
# Run with token
docker run -d \
-p 8001:8001 \
-e GITHUB_TOKEN='ghp_your_token' \
--name github-server \
--restart unless-stopped \
github-mcp-pro
# View logs
docker logs -f github-server
# Stop
docker stop github-serverDocker Compose
version: '3.8'
services:
github-mcp:
build: .
ports:
- "8001:8001"
environment:
- GITHUB_TOKEN=${GITHUB_TOKEN}
restart: unless-stopped
healthcheck:
test: ["CMD", "python", "-c", "import httpx; httpx.get('http://localhost:8001/mcp/v1/health')"]
interval: 30s
timeout: 10s
retries: 3______________________________________________________________________
☁️ 云部署
部署到铁路
# Install Railway CLI
npm install -g @railway/cli
# Login
railway login
# Initialize
railway init
# Deploy
railway up
# Set environment
railway variables set GITHUB_TOKEN=ghp_your_token
# Get URL
railway domain部署以渲染
- 连接您的GitHub存储库
- 创建新的Web服务
- 设置环境变量：
- GITHUB_TOKEN：您的代币
- 构建命令：
pip install -r requirements.txt - 启动命令：
python github_server.py - 部署！
部署到Fly.io
# Install flyctl
curl -L https://fly.io/install.sh | sh
# Login
flyctl auth login
# Launch
flyctl launch
# Set secret
flyctl secrets set GITHUB_TOKEN=ghp_your_token
# Deploy
flyctl deploy
# Check status
flyctl status部署到Heroku
# Create Procfile
echo "web: python github_server.py" > Procfile
# Create app
heroku create github-mcp-pro
# Set token
heroku config:set GITHUB_TOKEN=ghp_your_token
# Deploy
git push heroku main
# Check logs
heroku logs --tail______________________________________________________________________
🔍 高级搜索查询
GitHub支持强大的搜索语法：
# Search by stars
"query": "web framework stars:>10000"
# Search by language
"query": "machine learning language:python"
# Search by topic
"query": "topic:artificial-intelligence"
# Search by license
"query": "license:mit"
# Search in organization
"query": "org:google android"
# Search by size
"query": "size:>1000 language:go"
# Search by fork count
"query": "forks:>500 language:rust"
# Search by pushed date
"query": "pushed:>2024-01-01"
# Combined
"query": "neural network language:python stars:>1000 topic:deep-learning"______________________________________________________________________
📊 费率限制管理
检查你的极限
# Use the tool
get_rate_limit()
# Returns:
{
"core": {
"limit": 5000,
"remaining": 4850,
"reset": "2024-12-15T10:30:00"
},
"search": {
"limit": 30,
"remaining": 28,
"reset": "2024-12-15T09:45:00"
}
}最佳实践
- 使用认证 -5000次对60次请求/小时
- 缓存结果 -内置5分钟缓存
- 批量请求 -每次通话使用更高的限额
- 监控使用情况 -定期检查费率限制
- 处理错误 -优雅的退化
速率限制标头
服务器自动处理：
X-RateLimit-Limit-总限额X-RateLimit-Remaining-请求已离开X-RateLimit-Reset-重置时间戳Retry-After-等待几秒钟
______________________________________________________________________
🛡️ 安全最佳实践
许可证管理
# ✅ DO: Use environment variables
export GITHUB_TOKEN='ghp_...'
# ✅ DO: Use .env file (add to .gitignore)
echo "GITHUB_TOKEN=ghp_..." > .env
echo ".env" >> .gitignore
# ❌ DON'T: Hardcode in files
GITHUB_TOKEN = "ghp_..." # NEVER DO THIS!
# ❌ DON'T: Commit to git
git add .env # NEVER DO THIS!令牌范围
最低要求：
- ✅
public_repo-访问公共存储库 - ✅
read:user-读取用户资料
更多功能可选：
repo-访问私有存储库read:org-读取组织数据admin:repo_hook-管理webhooks
代币轮换
# Create new token
1. Visit https://github.com/settings/tokens
2. Generate new token
3. Update environment variable
4. Restart server
5. Revoke old token______________________________________________________________________
🧪 测试
手动测试
# Test repository info
curl -X POST http://localhost:8001/mcp/v1/tools/get_repository_info \
-H "Content-Type: application/json" \
-d '{"owner": "facebook", "repo": "react"}'
# Test search
curl -X POST http://localhost:8001/mcp/v1/tools/search_repositories \
-H "Content-Type: application/json" \
-d '{"query": "fastapi", "limit": 5}'
# Test user profile
curl -X POST http://localhost:8001/mcp/v1/tools/get_user_profile \
-H "Content-Type: application/json" \
-d '{"username": "torvalds"}'Python测试
import requests
# Test connection
response = requests.post(
"http://localhost:8001/mcp/v1/tools/server_info",
json={}
)
assert response.status_code == 200
print(response.json())
# Test repository
response = requests.post(
"http://localhost:8001/mcp/v1/tools/get_repository_info",
json={"owner": "python", "repo": "cpython"}
)
data = response.json()
assert "name" in data
print(f"Repository: {data['name']}")______________________________________________________________________
📈 监控和记录
服务器日志
服务器提供详细的日志记录：
[2024-12-15 10:30:45] INFO - Server started on port 8001
[2024-12-15 10:31:12] INFO - Tool called: get_repository_info
[2024-12-15 10:31:13] INFO - GitHub API request: repos/facebook/react
[2024-12-15 10:31:14] INFO - Response cached for 300s
[2024-12-15 10:31:14] INFO - Tool execution completed: 1.2s健康监测
# Check server health
curl http://localhost:8001/health
# Check rate limits
curl -X POST http://localhost:8001/mcp/v1/tools/get_rate_limit \
-H "Content-Type: application/json" \
-d '{}'______________________________________________________________________
🐛 故障排除
常见问题
“超出费率限制”
# Solution: Add GitHub token
export GITHUB_TOKEN='ghp_your_token'“未找到存储库”
# Check: Owner and repo spelling
# Check: Repository is public (or token has access)
# Check: Repository hasn't been deleted/renamed“连接被拒绝”
# Check: Server is running
python github_server.py
# Check: Port 8001 is available
lsof -i :8001 # Linux/Mac
netstat -ano | findstr :8001 # Windows“无效令牌”
# Solution: Regenerate token with correct scopes
# Required: public_repo, read:user“超时错误”
# Check: Internet connection
# Check: GitHub status: https://www.githubstatus.com/
# Increase: Timeout in code (REQUEST_TIMEOUT)调试模式
# Add to github_server.py for debugging
import logging
logging.basicConfig(level=logging.DEBUG)______________________________________________________________________
🎓 高级用法
自定义工具实现
添加您自己的工具 github_server.py:
@mcp.tool()
@safe_json_response
def your_custom_tool(param: str) -> dict:
"""
Your custom GitHub tool.
Args:
param: Parameter description
Returns:
Custom data
"""
data = gh_client.request(f"your/endpoint/{param}")
return {
"result": data
}批处理
# Process multiple repos
repos = [
("facebook", "react"),
("microsoft", "vscode"),
("google", "flutter")
]
for owner, repo in repos:
info = get_repository_info(owner, repo)
print(f"{owner}/{repo}: {info['statistics']['stars']} stars")缓存策略
服务器包括内置缓存（5分钟）。定制：
from functools import lru_cache
@lru_cache(maxsize=128)
def cached_request(endpoint: str):
return gh_client.request(endpoint)______________________________________________________________________
📋 需求
- python 3.10或更高
- 依赖关系：
- fastmcp>=2.14.0 - httpx>=0.24.0 - python-dotenv>=1.0.0
______________________________________________________________________
🤝 贡献
欢迎投稿！方法如下：
- 分叉 存储库
- 创建 特征分支
- 添加 您的改进
- 测试 彻底地
- 提交 pull请求
贡献的想法
- 添加更多GitHub功能
- 实现GraphQL API支持
- 添加webhook处理程序
- 创建可视化工具
- 添加分析仪表板
- 改进缓存策略
- 添加更多搜索筛选器
- 增强错误消息
- 添加速率限制预测
- 创建CLI界面
______________________________________________________________________
📄 许可证
MIT许可证-有关详细信息，请参阅许可证文件
______________________________________________________________________
🙏 学分
- GitHub API -数据提供者
- FastMCP -MCP框架
- Anthropic -Claude和MCP规范
- httpx -HTTP客户端
- 贡献者 非常感谢。
______________________________________________________________________
📞 支持
- 文档： 此自述文件
- 问题：
- 讨论：
- API文件：
______________________________________________________________________
⭐ 明星历史
如果这个项目对你有帮助，请考虑给它一个⭐!
______________________________________________________________________