{
  "id": "rubric:code-quality",
  "_kind": "Rubric",
  "_file": "benchmarks/eval-harnesses/rubrics.yaml",
  "_cluster": "benchmarks",
  "attributes": {
    "displayName": "Code-quality rubric",
    "scaleKind": "likert",
    "criteria": [
      {
        "name": "correctness",
        "description": "Does the code implement the intended behavior?",
        "scale": "1-5",
        "weight": 0.4
      },
      {
        "name": "readability",
        "description": "Is the code clear and well-named?",
        "scale": "1-5",
        "weight": 0.2
      },
      {
        "name": "idiomaticity",
        "description": "Idiomatic for the target language.",
        "scale": "1-5",
        "weight": 0.2
      },
      {
        "name": "performance",
        "description": "No obvious algorithmic issues.",
        "scale": "1-5",
        "weight": 0.2
      }
    ],
    "description": "Four-dimension code rubric for SWE-bench-style judging.\n"
  },
  "outgoingEdges": [],
  "incomingEdges": [
    {
      "from": "eval-run:gaia.claude-code.2025",
      "to": "rubric:code-quality",
      "kind": "scored_against_rubric",
      "attributes": {}
    }
  ]
}