id: '184'
narrative_anchor_date: '2028-01-01'
test: Please review the support-effort claim in `doc_helio_start_2028_authority_checklist_proposal` and
  post a correction comment that replaces the causal improvement framing with the certified mature authority-checklist
  cohort evidence and its appropriate limitation.
load_bearing_facts:
- 247
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_authority_checklist_correction
      path: result.ok
      value: true
      check_id: riley_184_00
    - check_id: riley_184_01
      type: field_equals
      tool: post_doc_comment
      action_id: post_authority_checklist_correction
      path: args.doc_id
      value: doc_helio_start_2028_authority_checklist_proposal
    - type: field_llm_judge
      tool: post_doc_comment
      action_id: post_authority_checklist_correction
      path: args
      criterion: 'The comment identifies the evidence as covering all 45 organic Helio Start workspaces
        that received the authority checklist from September 15 through October 31, and states that all
        45 have complete 90-day account-level support windows. Missing or contradictory cohort, date-window,
        denominator, or completeness information must fail.


        The comment reports support effort of 24 minutes at the median and 86 minutes at the 90th percentile,
        and states that 3 of 45 workspaces exceeded 90 support minutes. Missing or contradictory values
        must fail.


        The comment states that no unsupported case entered the admitted path. Missing or contradictory
        language must fail.


        The comment explicitly corrects or rejects the proposal''s causal support-improvement framing
        and says the cohort result is descriptive, not causal, because there was no randomized control.
        Generic caution without both the descriptive limitation and the lack of randomized control must
        fail.'
      check_id: riley_184_02
mock_state:
  docs:
  - id: doc_helio_start_2028_authority_checklist_proposal
    title: Helio Start 2028 authority-checklist proposal
    body: '# Helio Start 2028 authority-checklist proposal


      ## Evidence basis

      The earlier authority-checklist cohort proved that the checklist reduced support effort, so the
      improvement can be treated as an established benefit of the proposed approach.


      ## Proposal status

      Working draft for review.'
    folder: Growth
    created_at: '2028-01-01T09:00:00-06:00'