/evals/cases/using-agent-skills.json

{
  "skill_name": "using-agent-skills",
  "trigger": {
    "positive": [
      {
        "prompt": "Which skill should I use for this task?",
        "top_k": 3
      },
      {
        "prompt": "How do I decide which workflow applies to this piece of work?",
        "top_k": 3
      },
      {
        "prompt": "Route this request to the right skill in the pack",
        "top_k": 3
      }
    ],
    "negative": [
      {
        "prompt": "Debug the null pointer crash in checkout",
        "owner": "debugging-and-error-recovery"
      },
      {
        "prompt": "Make the modal accessible for keyboard users"
      }
    ]
  },
  "evals": [
    {
      "id": 1,
      "prompt": "A user asks: 'the login page is broken after yesterday's deploy'. Decide which skill applies and why.",
      "expected_output": "Correct routing through the decision tree with the chosen skill and rationale",
      "files": [
        "using-agent-skills"
      ],
      "expectations": [
        "The chosen skill matches the decision tree in the meta-skill",
        "The rationale references the routing logic rather than guessing",
        "Core operating behaviors (assumptions surfaced) are respected"
      ]
    }
  ]
}