{
  "context": "Tests whether the agent correctly identifies that bulk/PR-based optimization of multiple skills across external repositories should use `bulk-optimize` rather than the single-skill optimization workflow. The skill explicitly states that bulk-optimize is the correct tool for this scenario.",
  "type": "weighted_checklist",
  "checklist": [
    {
      "name": "bulk-optimize mentioned",
      "description": "Plan explicitly mentions `bulk-optimize` as the recommended tool for this type of task",
      "max_score": 35
    },
    {
      "name": "bulk-optimize for PR/repo work",
      "description": "Plan connects `bulk-optimize` to the multi-repository and/or pull request context (not just mentions it generically)",
      "max_score": 25
    },
    {
      "name": "Single-skill approach not recommended",
      "description": "Plan does NOT recommend iterating through all 40 files individually with the standard skill-optimizer workflow as the primary approach",
      "max_score": 20
    },
    {
      "name": "Rationale provided",
      "description": "Plan explains why bulk-optimize is preferred over individual skill optimization for this scenario (scale, PR workflow, or cross-repo considerations)",
      "max_score": 20
    }
  ]
}