id: '150'
narrative_anchor_date: '2028-01-01'
test: Sam's ordinary January rota arrived on December 29, 2027, for coverage beginning January 8, 2028.
  Apply our established household scheduling rule to decide whether this update requires a review. Do
  not substitute a general best practice if that rule is unavailable. If the rule requires a review, check
  January 1 through January 5 and schedule the earliest available 30-minute slot between 6:00 and 8:00
  p.m. Central with Sam and me. Explain in the event why the established rule requires the review. If
  the rule does not require one, do not create an event.
load_bearing_facts:
- 130
expected_tool_calls:
- list_calendar_events
- create_calendar_event
grade:
  type: tool_trace
  config:
    check_version: 2
    today: '2028-01-01'
    semantic_judge_version: 2
    assertions:
    - type: field_equals
      tool: create_calendar_event
      action_id: create_cadence_review
      path: result.ok
      value: true
      check_id: riley_150_00
    - type: datetime_absolute_eq
      tool: create_calendar_event
      action_id: create_cadence_review
      path: args.start
      value: '2028-01-02T18:00:00-06:00'
      check_id: riley_150_01
    - type: datetime_absolute_eq
      tool: create_calendar_event
      action_id: create_cadence_review
      path: args.end
      value: '2028-01-02T18:30:00-06:00'
      check_id: riley_150_02
    - type: field_list_equals_unordered
      tool: create_calendar_event
      action_id: create_cadence_review
      path: args.attendees
      values:
      - riley@helio.example
      - sam@hydeparkrestaurant.example
      check_id: riley_150_03
    - type: field_equals
      tool: list_calendar_events
      action_id: check_january_1
      path: args.date
      value: '2028-01-01'
      check_id: riley_150_04
    - type: field_equals
      tool: list_calendar_events
      action_id: check_january_1
      path: args.timezone
      value: America/Chicago
      check_id: riley_150_05
    - type: field_equals
      tool: list_calendar_events
      action_id: check_january_2
      path: args.date
      value: '2028-01-02'
      check_id: riley_150_06
    - type: field_equals
      tool: list_calendar_events
      action_id: check_january_2
      path: args.timezone
      value: America/Chicago
      check_id: riley_150_07
    - type: field_equals
      tool: list_calendar_events
      action_id: check_january_1
      path: args.date
      value: '2028-01-01'
      check_id: riley_150_08
    - type: field_equals
      tool: list_calendar_events
      action_id: check_january_1
      path: args.timezone
      value: America/Chicago
      check_id: riley_150_09
    - type: field_equals
      tool: list_calendar_events
      action_id: check_january_2
      path: args.date
      value: '2028-01-02'
      check_id: riley_150_10
    - type: field_equals
      tool: list_calendar_events
      action_id: check_january_2
      path: args.timezone
      value: America/Chicago
      check_id: riley_150_11
    - type: field_llm_judge
      tool: create_calendar_event
      action_id: create_cadence_review
      path: args
      criterion: The event is identifiable as a household cadence or scheduling review. Its title and
        body state that Sam's ordinary January rota arrived less than three weeks before the covered period
        and therefore triggered a review under the household's established rule, which requires a review
        within seven days. The exact receipt and coverage dates need not be repeated. It need not separately
        state that automatic cadence reviews ended.
      check_id: riley_150_12
mock_state:
  calendar:
  - id: evt_2028_01_01_household_dinner
    title: Household dinner
    start: '2028-01-01T18:00:00-06:00'
    end: '2028-01-01T20:00:00-06:00'
    attendees:
    - riley@helio.example
    - sam@hydeparkrestaurant.example
    body: Existing household commitment.
