Cleanup

The model receives the prompt (and optional system message). The run uses scorer contains_any with the JSON configuration below. Pass/fail and partial credit are determined entirely by that scorer against the model output; no human grading.

Refactor without changing behavior. Return only Python code.

def active_ids_04(users):
    out = []
    for user in users:
        if user.get('active') == True:
            out.append(user['id'])
    return out

{
  "expected_contains": [
    "for user in users",
    "user.get('active')",
    "out.append"
  ]
}