id: '112'
narrative_anchor_date: '2028-01-01'
test: Review the newest benefits message in my inbox and create a January reconciliation note. For each
  topic, first state the existing 2028 baseline, then state exactly what the new message confirms or does
  not confirm. Cover the plan, coverage period, payroll deductions, cost limits, provider-network result,
  and employer HSA schedule and conditions.
load_bearing_facts:
- 282
- 283
- 284
- 296
expected_tool_calls:
- list_inbox
- 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: reconciliation_doc
      path: result.ok
      value: true
      check_id: riley_112_00
    - type: field_equals
      tool: list_inbox
      action_id: inbox_review
      path: args.folder
      value: inbox
      check_id: riley_112_01
    - type: field_llm_judge
      tool: create_doc
      action_id: reconciliation_doc
      path: args
      criterion: 'The document identifies the election baseline as the Helio 2028 Employee-Only HDHP and
        states that its coverage period is January 1 through December 31, 2028.


        The document accurately states the payroll structure: $98 deducted from each of 24 paychecks,
        totaling $2,352 for the year.


        The document accurately states a $2,000 deductible and a $4,000 out-of-pocket maximum.


        The document records that the primary-care clinician, physical-therapy clinic, and eye-care clinic
        were confirmed to be in the final network when the election was made.


        The document accurately states that Helio''s employer HSA funding consists of four $250 deposits
        scheduled for January 15, April 15, July 15, and October 15, 2028, and clearly states that each
        individual deposit depends on remaining HSA-eligible on its deposit date.


        The document clearly distinguishes the new inbox-visible confirmation from the historical election
        baseline. It reports that the newest benefits message visibly confirms that 2028 medical coverage
        is now active and treats that as consistent with the expected January 1 start. It also explicitly
        flags that the message does not itself confirm the specific HDHP election or full coverage period,
        payroll amounts and paycheck count, deductible and out-of-pocket maximum, provider-network status,
        or HSA deposit amounts, dates, and eligibility conditions. It must not present those omitted details
        as newly confirmed by the email, and it should indicate that no actual contradiction is visible
        from the limited confirmation.'
      check_id: riley_112_02
mock_state:
  inbox:
  - id: helio-benefits-2028-activation
    from: Helio Benefits <benefits@helio.example>
    subject: Your 2028 medical coverage is now active
    date: '2028-01-01T07:15:00-05:00'
    category: Benefits