2. field_llm_judge / create_doc
{
"type": "field_llm_judge",
"tool": "create_doc",
"action_id": "create_hypothesis_card",
"path": "args",
"criterion": "The created document is for EXP-2023-02-pricing-page and explicitly labels its content as the original design hypothesis rather than an observed result. It states that the treatment concept was a simplified layout plus explicit value-per-seat framing, targeted the starter and growth customer tiers, and was intended to increase conversion by at least 1.5 percentage points. Equivalent wording is acceptable, but 1.5 points must be presented as the minimum intended lift, not as measured performance.",
"check_id": "riley_026_01"
}Complete grading specification
{
"type": "tool_trace",
"config": {
"check_version": 2,
"today": "2028-01-01",
"semantic_judge_version": 2,
"assertions": [
{
"type": "field_equals",
"tool": "create_doc",
"action_id": "create_hypothesis_card",
"path": "result.ok",
"value": true,
"check_id": "riley_026_00"
},
{
"type": "field_llm_judge",
"tool": "create_doc",
"action_id": "create_hypothesis_card",
"path": "args",
"criterion": "The created document is for EXP-2023-02-pricing-page and explicitly labels its content as the original design hypothesis rather than an observed result. It states that the treatment concept was a simplified layout plus explicit value-per-seat framing, targeted the starter and growth customer tiers, and was intended to increase conversion by at least 1.5 percentage points. Equivalent wording is acceptable, but 1.5 points must be presented as the minimum intended lift, not as measured performance.",
"check_id": "riley_026_01"
}
]
}
}