{
  "$type": "site.standard.document",
  "bskyPostRef": {
    "cid": "bafyreifl4jtw6ztu3b5y2kl4buv6pl6j5rt7dd6rhz75izsuhbcavtwsqi",
    "uri": "at://did:plc:pgryn3ephfd2xgft23qokfzt/app.bsky.feed.post/3miwvrvfnvor2"
  },
  "path": "/t/looking-for-simple-ways-to-evaluate-an-ai-agent/175062#post_1",
  "publishedAt": "2026-04-07T19:22:10.000Z",
  "site": "https://discuss.huggingface.co",
  "tags": [
    "test.qlankr.com"
  ],
  "textContent": "I’m building an AI agent that answers questions based on documentation/small knowledge base and I’m trying to figure out a simple way to evaluate if it is working well.\n\nI have used test.qlankr.com, which looks interesting, but I’m wondering if there are any other eval tools people here use that are beginner-friendly and make it easy to share results clearly.\n\nWhat I’m mainly looking for is something that helps with:\n\n  * comparing outputs\n  * seeing weak points or regressions\n  * seeing where the agent gives incomplete orbad\n  * sharing result with other people without making it look too complicated\n\n\n\nCurious to find out what people here are using for this.\n\nCheers",
  "title": "Looking for simple ways to evaluate an AI agent"
}