id: 089
narrative_anchor_date: '2028-01-01'
test: Create a compact historical note covering the first two hypothesis rulings in the February 2023
  mid-segment churn investigation. Identify H1 and H2, who made each ruling, their dates and reasons,
  and the hypotheses that remained after H2. Use people's names rather than `the first-person speaker`
  or another role label. For H2, distinguish the person who pulled the comparisons and made the ruling
  from Owen, who performed that morning's check.
load_bearing_facts:
- 309
- 310
- 311
- 317
- 323
- 324
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_first_rulings_note
      path: result.ok
      value: true
      check_id: riley_089_00
    - type: field_llm_judge
      tool: create_doc
      action_id: create_first_rulings_note
      path: args
      criterion: The note identifies H1 as the pricing-artifact hypothesis and says Owen closed it as
        out for now on February 15, 2023. It identifies H2 as competitor cycle or a generic seasonal effect
        and says Riley ruled it out on February 21 because historical February comparisons showed no matching
        spike. It says H3 onboarding timing and H4 Customer Success coverage remained after H2.
      check_id: riley_089_01
mock_state: {}