ai-agents-test-case-creation-after-writing-a-skill-draft-before-running-evaluat-2284206d

THEN Create 2-3 realistic test prompts that a user would actually say. Save them in `evals/evals.json` with the skill name and test cases (prompt, expected_output, files). Do not include assertions yet; they will be drafted later while runs are in progress.

```json
{
  "skill_name": "example-skill",
  "evals": [
    {
      "id": 1,
      "prompt": "User's task prompt",
      "expected_output": "Description of expected result",
      "files": []
    }
  ]
}
```