Community

LLM Eval Harness

io.github.razilzul/llm-eval-harness

Run a golden set and gate LLM model swaps on accuracy, with latency and cost per request.

Install

Add the config to your client settings. Replace values in <angle brackets> with your own.

claude_desktop_config.json

{
  "mcpServers": {
    "llm-eval-harness": {
      "command": "docker",
      "args": [
        "run",
        "-i",
        "--rm",
        "-e",
        "ANTHROPIC_API_KEY",
        "-e",
        "GEMINI_API_KEY",
        "ghcr.io/razilzul/llm-eval-harness:1.0.0"
      ],
      "env": {
        "ANTHROPIC_API_KEY": "<ANTHROPIC_API_KEY>",
        "GEMINI_API_KEY": "<GEMINI_API_KEY>"
      }
    }
  }
}

Details

Name: io.github.razilzul/llm-eval-harness
Version: 1.0.0
Updated: Oct 18, 2026
Source: Official MCP Registry
Repository: https://github.com/razilzul/llm-eval-harness