License Scanner MCP Server
Enables scanning of project dependencies across multiple package managers (npm, pip, cargo, etc.) and generates comprehensive markdown license reports. Supports automatic license detection from package registries with caching for improved performance.
README
License Scanner MCP Server
A Model Context Protocol (MCP) server built with FastMCP that scans project dependencies and generates license reports in markdown format.
Features
- Multi-package manager support: npm, pip, cargo, and more
- Automatic license detection: Fetches license information from package registries
- Markdown report generation: Creates comprehensive license reports
- Caching: Avoids repeated API calls for better performance
- Error handling: Robust error handling and informative error messages
Supported Package Managers
- npm/yarn/pnpm:
package.json,yarn.lock,pnpm-lock.yaml - Python:
requirements.txt,pyproject.toml,Pipfile - Rust:
Cargo.toml,Cargo.lock - PHP:
composer.json,composer.lock - Ruby:
Gemfile,Gemfile.lock - Go:
go.mod,go.sum - Java:
pom.xml,build.gradle
Installation
- Install dependencies:
pip install -r requirements.txt
- Run the MCP server:
python license_scanner.py
Available Tools
1. scan_dependencies(project_path: str)
Scans a project for dependencies and returns detailed license information.
Parameters:
project_path: Path to the project directory to scan
Returns: JSON string containing dependency and license information
2. generate_license_report(project_path: str, output_file: str = None)
Generates a markdown license report for a project.
Parameters:
project_path: Path to the project directory to scanoutput_file: Optional path to save the markdown report
Returns: Markdown content of the license report
3. list_package_managers(project_path: str)
Lists all package manager files found in a project.
Parameters:
project_path: Path to the project directory to scan
Returns: JSON string containing list of package manager files
Example Usage
Using the MCP server with Claude Desktop
- Add the server to your Claude Desktop configuration:
{
"mcpServers": {
"license-scanner": {
"command": "python",
"args": ["/path/to/license_scanner.py"],
"cwd": "/path/to/project"
}
}
}
- Use the tools in Claude Desktop:
- "Scan the dependencies in my project"
- "Generate a license report for this project"
- "List the package managers used in this project"
Example Project
The example_project/ directory contains sample files for different package managers:
package.json- npm dependenciesrequirements.txt- Python dependenciesCargo.toml- Rust dependencies
License Information Sources
- npm packages: npm registry API
- Python packages: PyPI JSON API
- Rust crates: crates.io API
Output Format
The license report includes:
- Project information: Path, total dependencies, package files found
- Dependencies by manager: Organized by package manager with license details
- License summary: Count of dependencies by license type
Sample Markdown Output
# License Report
**Project:** /path/to/project
**Total Dependencies:** 8
**Package Files Found:** package.json, requirements.txt
## NPM Dependencies
Found 5 dependencies
| Package | Version | License | Author | Homepage |
|---------|---------|---------|--------|----------|
| express | ^4.18.2 | MIT | TJ Holowaychuk | https://expressjs.com/ |
| lodash | ^4.17.21 | MIT | John-David Dalton | https://lodash.com/ |
## License Summary
| License | Count |
|---------|-------|
| MIT | 6 |
| Apache-2.0 | 2 |
Error Handling
The server includes comprehensive error handling:
- Invalid project paths
- Missing package files
- Network errors when fetching license information
- Malformed package files
Performance
- Caching: License information is cached to avoid repeated API calls
- Parallel processing: Multiple package files are processed efficiently
- Timeout handling: API requests have timeout protection
Contributing
- Fork the repository
- Create a feature branch
- Add tests for new functionality
- Submit a pull request
License
This project is licensed under the MIT License.
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。