[
  {
    "type": "noul",
    "statement": "The MCP tools, resources and prompts the agent called (`mcp_tools_called`, `mcp_resources_called`, `mcp_prompts_called`, or `tools_called`) are the right ones among those in `mcp_servers` for what `input` asks, with no clearly better primitive missed and no unnecessary call.",
    "weight": 2
  },
  {
    "type": "noul",
    "statement": "Every MCP call the agent made passes arguments that match that primitive's input schema in `mcp_servers` and carry the values `input` calls for."
  },
  {
    "type": "score",
    "question": "Overall, how well does the agent use the MCP servers in `mcp_servers`, in both the primitives chosen and the arguments passed, to address `input`?",
    "levels": ["Wrong use", "Poor use", "Adequate use", "Correct use"]
  }
]
