SCANOSS MCP Server

SCANOSS MCP Server

A Model Context Protocol (MCP) server that provides tools for scanning code for open source license compliance using SCANOSS technology.

Overview

The SCANOSS MCP Server integrates SCANOSS's open source scanning capabilities with the Model Context Protocol, allowing Language Models to:

1. Scan code snippets for undeclared open source content
2. Check license compliance for combinations of open source licenses
3. Get actionable insights on open source components in code

By leveraging this server, AI assistants can provide real-time feedback on code they generate or modify, ensuring compliance with open source licenses and reducing legal risks.

Features

• Code Scanning: Detect open source components in code snippets, including full components, files, and snippets.
• License Compliance: Analyze combinations of licenses for compatibility issues and compliance requirements.
• Formatted Results: Receive well-formatted, actionable insights on scan results.
• Optional API Key: Works with both free (osskb.org) and enterprise SCANOSS instances.

Requirements

• Python 3.12+
• uv for Python package management
• SCANOSS Python SDK
• MCP Python SDK

Installation

Local Installation

1. Clone this repository:

   git clone https://github.com/matiasdaloia/scanoss-mcp-server.git
   cd scanoss-mcp-server
   

2. Install dependencies using uv:

   uv venv
   uv pip install -e .
   

3. (Optional) Create a .env file in the root directory with your SCANOSS API key:

   SCANOSS_API_KEY=your_api_key_here
   SCANOSS_API_URL=https://your-scanoss-instance.com/api/scan/direct
   

   If no API key is provided, the server will use the free osskb.org service.

Docker Installation

1. Pull the pre-built Docker image:

   docker pull mcp/scanoss-mcp-server
   

   Or build it yourself:

   docker build -t mcp/scanoss-mcp-server .
   

2. Run using Docker:

   docker run -i --rm -e SCANOSS_API_KEY=your_key_here -e SCANOSS_API_URL=https://your-scanoss-instance.com/api/scan/direct mcp/scanoss-mcp-server
   

Usage

Running the Server

Locally with uv

Run the MCP server:

uv run scanoss_mcp_server.py

With Docker

docker run -i --rm -e SCANOSS_API_KEY=your_key_here mcp/scanoss-mcp-server

Client Configuration

To use the SCANOSS MCP Server with your AI assistant or MCP client, you'll need to configure it properly:

For Claude Desktop

Add the following to your Claude Desktop configuration file (claude_desktop_config.json):

"scanoss-mcp-server": {
  "command": "uv",
  "args": [
    "run",
    "scanoss_mcp_server.py"
  ],
  "env": {
    "SCANOSS_API_KEY": "your_api_key_here",
    "SCANOSS_API_URL": "https://your-scanoss-instance.com/api/scan/direct"
  }
}

Or if using Docker:

"scanoss-mcp-server": {
  "command": "docker",
  "args": [
    "run",
    "-i",
    "--rm",
    "-e", "SCANOSS_API_KEY",
    "-e", "SCANOSS_API_URL",
    "mcp/scanoss-mcp-server"
  ],
  "env": {
    "SCANOSS_API_KEY": "your_api_key_here",
    "SCANOSS_API_URL": "https://your-scanoss-instance.com/api/scan/direct"
  }
}

Available Tools

The server exposes the following tools to MCP clients:

1. scan_code_snippet: Scans a code snippet for open source components
   • Parameters:
     • code: The code snippet to scan (required)
     • file_name: Name of the file (required)

Contributing

Contributions are welcome! Please feel free to submit a Pull Request.

License

This project is licensed under the MIT License - see the LICENSE file for details.

Acknowledgments

• SCANOSS for their open source scanning technology
• Model Context Protocol for the MCP standard and SDK
• uv for fast Python package management