{
  "title": "AI Eval Plan Generator",
  "description": "Turns a feature description into a concrete evaluation plan: dimensions to score, a rubric, and a minimum viable test set.",
  "category": "AI Evaluation and Testing",
  "tags": [
    "evaluation",
    "testing",
    "planning"
  ],
  "templateContent": "You are an AI evaluation designer. Produce an evaluation plan for the following feature.\n\nFeature description:\n{{feature_description}}\n\nPrimary risk to guard against: {{primary_risk}}\n\nProduce:\n1. 3-6 evaluation dimensions relevant to this feature (e.g. correctness, safety, tone, latency-sensitivity).\n2. A 1-5 scoring rubric for each dimension, with a one-sentence description of each score level.\n3. A minimum viable test set: {{test_case_count}} concrete test cases (inputs + what a passing output looks like), covering typical use, edge cases, and the primary risk above.\n4. A pass/fail threshold recommendation for a release gate.\n\nOutput format: {{output_format}}.",
  "variables": [
    {
      "name": "feature_description",
      "label": "Feature description",
      "type": "multiline",
      "required": true,
      "default": null,
      "allowedValues": null
    },
    {
      "name": "primary_risk",
      "label": "Primary risk to guard against",
      "type": "text",
      "required": true,
      "default": null,
      "allowedValues": null
    },
    {
      "name": "test_case_count",
      "label": "Number of test cases",
      "type": "number",
      "required": true,
      "default": 10,
      "allowedValues": null
    },
    {
      "name": "output_format",
      "label": "Output format",
      "type": "select",
      "required": true,
      "default": "Markdown",
      "allowedValues": [
        "Markdown",
        "JSON",
        "Plain text"
      ]
    }
  ],
  "usageInstructions": "Describe the feature in plain language, name the one failure mode you're most worried about, and run this before your first release and again after any prompt or model change.",
  "limitations": "The generated test cases are a starting set, not a complete suite; domain experts should review them before relying on the plan for a real release decision."
}