id: '140'
narrative_anchor_date: '2028-01-01'
test: Please review the draft executive-communications retrospective at document ID `doc_1830355200001`
  and post a comment correcting its account of the April 3, 2023 onboarding-trigger briefing. The correction
  should reflect the stance held at that checkpoint and the additional confirmation still needed before
  describing the result as fully settled. Do not edit the document itself.
load_bearing_facts:
- 355
expected_tool_calls:
- post_doc_comment
grade:
  type: tool_trace
  config:
    check_version: 2
    today: '2028-01-01'
    semantic_judge_version: 2
    assertions:
    - type: field_equals
      tool: post_doc_comment
      action_id: post_historical_correction
      path: result.ok
      value: true
      check_id: riley_140_00
    - check_id: riley_140_01
      type: field_equals
      tool: post_doc_comment
      action_id: post_historical_correction
      path: args.doc_id
      value: doc_1830355200001
    - type: field_llm_judge
      tool: post_doc_comment
      action_id: post_historical_correction
      path: args
      criterion: The comment must correct the draft's claim that the result was fully settled on April
        3, 2023. It must convey that Riley's contemporaneous executive framing was cautiously optimistic
        rather than conclusive and that Riley still wanted one additional post-ship cohort before treating
        the result as fully settled. Missing either the cautious framing or the specific post-ship-cohort
        confirmation boundary, or endorsing the draft's settled-win framing, must fail.
      check_id: riley_140_02
mock_state:
  docs:
  - id: doc_1830355200001
    title: Executive communications retrospective — onboarding-trigger briefing
    body: '# Executive communications retrospective


      ## April 3, 2023 onboarding-trigger briefing

      The second-cohort result made the onboarding-trigger experiment a settled win. The executive briefing
      could therefore present the result as conclusively established, with no further confirmation needed
      before leadership treated the matter as closed.'
    folder: Growth
    created_at: '2028-01-01T09:00:00-06:00'
