evalgate-mcp

HolisticShort/evalgate-mcp

★ 0 stars JavaScript Code/Dev Tools Updated today

MCP server exposing evalgate — run a suite, explain a failure, check drift, and see whether the judge was ever calibrated.

View on GitHub → 🔍 Audit Wallet Slippage →

Quick Install

Copy the config for your editor. Some servers may need additional setup — check the README.

Add to claude_desktop_config.json:

{
  "mcpServers": {
    "evalgate-mcp": {
      "command": "npx",
      "args": [
        "-y",
        "HolisticShort/evalgate-mcp"
      ]
    }
  }
}

README Excerpt

**An MCP server that lets an agent read its own eval results.** Wraps and ask whether the judge was ever calibrated. > **Phase 1 is read-only.** Nothing here writes a file. `run_suite` executes the system under test and may > bill a provider, but the result comes back in memory rather than landing in `.evalgate/result.json`.

Tools (7)

calibratedcheck_driftexplain_resultjudge_statuslist_suitesrun_suiteuncalibrated

evalgate-mcp

Quick Install

README Excerpt

Tools (7)

Related Code/Dev Tools Servers