GitLab PR 分析 MCP 服务器
该项目提供了一个 MCP（模型控制协议）服务器，它将 GitLab 合并请求分析与 Confluence 文档集成在一起。它允许您获取合并请求详细信息、分析代码更改并将结果存储在 Confluence 页面中。
特征
- 从 GitLab 获取合并请求详细信息
- 分析合并请求中的代码更改
- 生成详细报告，包括：
- 基本合并请求信息
- 代码变更统计
- 文件类型分析
- 详细文件变更
- 将分析结果存储在 Confluence 中
- 全面的调试日志记录
先决条件
- Python 3.8 或更高版本
- 具有 API 访问权限的 GitLab 帐户
- Confluence 帐户（可选，用于存储分析结果）
- 访问所需的 GitLab 项目
安装
- 克隆存储库：
git https://github.com/CodeByWaqas/MRConfluenceLinker-mcp-server.git
cd MRConfluenceLinker-mcp-server
- 创建并激活虚拟环境：
python -m venv .venv
source .venv/bin/activate # On Windows, use: .venv\Scripts\activate
- 安装依赖项：
pip install -r requirements.txt
或者
uv add "mcp[cli]" python-gitlab python-dotenv atlassian-python-api requests
配置
- 复制示例环境文件：
- 使用您的凭据编辑
.env文件：
GITLAB_URL=https://gitlab.com
GITLAB_TOKEN=your_gitlab_token
GITLAB_PROJECT_ID=your_project_id
# Optional Confluence integration
CONFLUENCE_URL=your_confluence_url
CONFLUENCE_USERNAME=your_username
CONFLUENCE_TOKEN=your_confluence_token
CONFLUENCE_SPACE=your_space_key
获取凭证
- GitLab Token ：在 GitLab 中生成具有
api范围的个人访问令牌 - Confluence Token ：在您的 Atlassian 帐户设置中生成 API 令牌
用法
- 启动 MCP 服务器：
python src/MRConfluenceLinker-mcp-server/server.py
或者
使用 Claude Desktop 进行设置
# claude_desktop_config.json
# Can find location through:
# Claude -> Settings -> Developer -> Edit Config
{
"mcpServers": {
"MRConfluenceLinker-mcp-server": {
"command": "uv",
"args": [
"--directory",
"/<Absolute-path-to-folder>/MRConfluenceLinker-mcp-server/src/MRConfluenceLinker-mcp-server",
"run",
"server.py"
]
}
}
}
2. The server will listen for commands through stdin/stdout. You can interact with it using prompts like:
您能从项目“my-project”获取合并请求 #1 的详细信息吗？您能分析项目“my-project”合并请求 #1 中的代码更改吗？您能将项目“my-project”合并请求 #1 的摘要存储在 Confluence 中吗？
## Available Tools
The server provides the following tools:
1. `fetch_mr_details`: Fetches details of a specific merge request or all merge requests
- Parameters:
- `project_id`: The GitLab project ID
- `mr_id` (optional): Specific merge request ID
2. `analyze_code_changes`: Analyzes code changes in a merge request
- Parameters:
- `project_id`: The GitLab project ID
- `mr_id`: The merge request ID to analyze
3. `store_in_confluence`: Stores analysis results in Confluence
- Parameters:
- `project_id`: The GitLab project ID
- `mr_id` (optional): Specific merge request ID
- `analysis` (optional): Analysis results to store
## Logging
The server generates detailed logs in `mcp_server.log` and outputs to stderr. This helps in debugging issues with:
- GitLab API access
- Confluence integration
- Code analysis
- Page creation and updates
## Error Handling
The server includes comprehensive error handling for:
- Missing environment variables
- API authentication issues
- Network connectivity problems
- Invalid project or merge request IDs
- Confluence permission issues
## Contributing
1. Fork the repository
2. Create a feature branch
3. Commit your changes
4. Push to the branch
5. Create a Pull Request
## License
This project is licensed under the MIT License - see the LICENSE file for details.
## Support
For support, please [create an issue](https://github.com/CodeByWaqas/MRConfluenceLinker-mcp-server/issues) or contact the maintainers.
## Project Structure
MRConfluenceLinker-mcp-server/ │ src/ # 源代码目录│ └── MRConfluenceLinker- mcp -server/ # 主服务器包 │ ...
### Key Components
- **Source Code**: Located in the `src/MRConfluenceLinker-mcp-server/` directory
- `server.py`: Main MCP server implementation
- `resources/client.py`: Client-side implementation contains GitLab PR integration
- **Configuration Files**:
- `requirements.txt`: Python package dependencies
- `pyproject.toml`: Project metadata and build configuration
- `uv.lock`: Locked dependency versions
- `.env.example`: Environment variables template
- **Documentation**:
- `README.md`: Project overview and setup instructions
- `CONTRIBUTING.md`: Contribution guidelines
- `LICENSE`: Project license
- **Development**:
- `__pycache__/`: Python cache files
- `mcp_server.log`: Server logs for debugging