skills/ai-slides/_scores.json

{
  "version": "1.0.0",
  "skillHash": "sha256:5e2304e0cd58f6f38661e875e00fd693bff90ec0ee439f2cec093db4d9feaca4",
  "scoredAt": "2026-05-13T13:46:36.389Z",
  "backend": "ollama",
  "model": "gpt-oss:20b",
  "quality": {
    "score": 93,
    "dimensions": {
      "clarity": "PASS",
      "completeness": "WEAK",
      "conciseness": "PASS",
      "actionability": "PASS",
      "crossPlatform": "PASS",
      "examples": "PASS"
    },
    "issues": [
      {
        "severity": "MEDIUM",
        "category": "completeness",
        "detail": "The instructions do not cover error scenarios or edge cases, focusing only on the happy path."
      }
    ]
  },
  "security": {
    "verdict": "SAFE",
    "issues": []
  },
  "impact": {
    "multiplier": 1.3,
    "baselineAvg": 33,
    "treatmentAvg": 43,
    "scenarios": [
      {
        "name": "pitch-deck-for-enterprise-ai-tool",
        "baseline": 35,
        "treatment": 15,
        "rationale": "Response A covers more of the required slides and includes key elements like market size, traction, and funding ask, whereas Response B only provides a few slides and omits many rubric requirements."
      },
      {
        "name": "q4-engineering-review-deck",
        "baseline": 30,
        "treatment": 70,
        "rationale": "Response A covers most rubric items but violates design limits; Response B lacks required slides, notes, and visuals."
      }
    ]
  }
}