id: '141'
narrative_anchor_date: '2028-01-01'
test: Please post a narrowly scoped correction comment on the archived April 21, 2023 staff note at document
  ID `doc_staff_archive_2023_04_21`. Correct the status language based on what I said at that meeting
  and identify the evidence gate that was still outstanding then. Keep the comment explicitly tied to
  the April 21 meeting rather than later outcomes.
load_bearing_facts:
- 372
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: comment_apr21_status
      path: result.ok
      value: true
      check_id: riley_141_00
    - check_id: riley_141_01
      type: field_equals
      tool: post_doc_comment
      action_id: comment_apr21_status
      path: args.doc_id
      value: doc_staff_archive_2023_04_21
    - type: field_llm_judge
      tool: post_doc_comment
      action_id: comment_apr21_status
      path: args
      criterion: 'The comment must be explicitly framed as a correction to the April 21, 2023 staff-meeting
        record and convey that, as of that meeting, the churn issue should not yet have been described
        as fixed: the leading hypothesis still appeared to hold, but cohort 3 still needed to land before
        Riley would call it fixed. Later outcomes must not be substituted for the status as of April 21.
        Missing or contradicting any of these points fails.'
      check_id: riley_141_02
mock_state:
  docs:
  - id: doc_staff_archive_2023_04_21
    title: Staff notes archive — April 21, 2023
    body: '## Mid-segment churn update

      The churn problem was declared fixed at today''s staff meeting following the onboarding-trigger
      rollout.'
    folder: Growth archive
    created_at: '2023-04-21T12:00:00-05:00'
