Score

Write the rubric#

Criteria is an array ordered from the first level to the last. Indices start at zero. Draft 0.1 requires at least two levels; providers may impose an upper limit.

1{
2  "type": "score",
3  "instructions": "How much does this affect the reporting workflow?",
4  "criteria": [
5    "No impact",
6    "Delayed work",
7    "Work blocked"
8  ]
9}

Interpret the score#

1{
2  "type": "score",
3  "score": 1.75,
4  "legend": {
5    "0": "No impact",
6    "1": "Delayed work",
7    "2": "Work blocked"
8  },
9  "probabilities": {
10    "0": 0.05,
11    "1": 0.15,
12    "2": 0.8
13  },
14  "confidence": 0.625
15}

The score is the expected level index: sum each index multiplied by its probability. For this example, 0 × 0.05 + 1 × 0.15 + 2 × 0.80 = 1.75. A score can fall between levels.

Legend maps the index strings back to level descriptions. Probabilities uses the same keys. Confidence is reported separately.

Keep the scale meaningful#

Rate one property per question. Mixing urgency, severity, and sentiment in one rubric makes the resulting number difficult to interpret. If you need all three, ask three questions.