MCP Wegene Assistant

🚀 WeGene Assistant MCP Server

An MCP server for WeGene Assistant that uses Large Language Models (LLMs) to analyze users' WeGene genetic testing reports.

✨ Features

Resources

Once users complete authorization, all reports under their accounts will be exposed as resources:

• Access each individual report using the custom wegene:// URI scheme.
• Each report resource includes a name, description, and a content type in application/json format.

Tools

The server implements the following tools:

• wegene-oauth: Initiate the WeGene Open API OAuth process in the browser.
  • Users should complete authorization within 120 seconds for the LLM to further access the reports.
• wegene-get-profiles: Read the list of profiles under the user's WeGene account.
  • Return the names and IDs of profiles for the LLM to use.
• wegene-get-report-info: Return report metadata so that the LLM knows the available reports.
  • Return a list of information such as report names, descriptions, and endpoints.
• wegene-get-report: Read the result of a single report under a specific profile.
  • Return the JSON result of the report specified by the WeGene Open API platform.
  • Parameters:
    • report_endpoint: The endpoint from which to obtain the report.
    • report_id: The ID of the report to obtain.
    • profile_id: The profile from which to obtain the report.

📦 Installation

Installation via Smithery

To automatically install the WeGene Assistant for the Claude desktop version via Smithery:

npx -y @smithery/cli install @xraywu/mcp-wegene-assistant --client claude

Local Installation

Prepare the MCP Server

1. Clone this project.
2. Run uv sync --dev --all-extras in the root folder of the project.

Configure the Claude Desktop Version

• For MacOS: ~/Library/Application\ 支持/Claude/claude_desktop_config.json
• For Windows: %APPDATA%/Claude/claude_desktop_config.json

Add the following content to the configuration file:

{
  "mcpServers": {
    "wegene-assistant": {
      "command": "uv",
      "args": [
        "--directory",
        "/path/to/wegene-assistant",
        "run",
        "wegene-assistant"
      ]
    }
  }
}

🔧 Configuration

• You need the WeGene Open API key/secret to use this project.
• Copy .env.example and name it .env, then update the key and secret in the file.