evalgate for Claude Desktop

io.github.ipezygj/evalgate

Statistical checks an agent runs before trusting an AI eval number (is #1 real, judge bias, more).

client:Claude Desktop transport:stdio runtime:pypi

Install evalgate in Claude Desktop

~/Library/Application Support/Claude/claude_desktop_config.json · windows: %APPDATA%\Claude\claude_desktop_config.json

{
  "mcpServers": {
    "evalgate": {
      "command": "uvx",
      "args": [
        "eval-integrity"
      ]
    }
  }
}

Settings → Developer → Edit Config opens this file. Restart the app after saving. Claude Desktop docs