id: '108'
narrative_anchor_date: '2028-01-01'
test: Create a compact decision-evidence sheet for the restaurant archive covering the three roasted-chicken
  sale runs. State the evidence-based conclusions about production and customer demand, and keep the sheet
  focused on the run evidence rather than making or reopening a menu decision.
load_bearing_facts:
- 47
expected_tool_calls:
- create_doc
grade:
  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_roasted_chicken_evidence_sheet
      path: result.ok
      value: true
      check_id: riley_108_00
    - type: field_llm_judge
      tool: create_doc
      action_id: create_roasted_chicken_evidence_sheet
      path: args
      criterion: 'The body is a compact decision-evidence sheet about the three roasted-chicken sale runs,
        rather than unrelated notes or a general menu summary.


        The body clearly states that across all three runs, production was repeatable and stayed within
        the chef-owner''s labor and yield thresholds. Missing or contradicting either repeatability or
        compliance with both types of threshold must fail.


        The body clearly states that customer demand or sell-through varied across the runs. Any claim
        that demand or sell-through was uniform must fail.


        The body does not make, recommend, or purport to reopen a decision about permanent, standing,
        weekly, private-event, or other future menu placement; it remains an evidence record of the sale
        runs.'
      check_id: riley_108_01
mock_state: {}