Test 129

Jan 1, 2028 / 1 fact

Request

Create a short decision record titled `Expansion Assist second cohort — eligibility integrity decision` for the cohort-governance archive. Explain how the operational capacity allowance related to the locked evidence-complete roster, record the decision on whether eligibility would be relaxed to use the remaining capacity, and make clear how the unused headroom should be interpreted.

Required memory

Fact 59

Riley Tanaka locked the reviewed mid-segment-account cohort at the 15 evidence-complete candidates rather than filling the approved 18-account cap with marginal records.

Source evidence (1)

002026Oct 5, 2024 / 09:40 UTC-05:00

Owen and Customer Success completed the candidate pass across 31 retained mid-segment accounts. Fifteen satisfy deeper integration use, a currently active qualifying human admin, and a product-visible usage ceiling crossed within the prior 14 days. Five are suppressed for active adoption or renewal work, six have an older ceiling event, three lack a currently active qualifying admin, and two lack sufficient integration depth. I am locking the cohort at those 15 evidence-complete candidates rather than filling the approved 18-account cap with marginal records. Post the lock and exclusion counts to `#growth-ops`; preserve the three-per-week ceiling and same-day active-admin and Customer Success checks before exposure.

Message 002026 in history

Expected tool calls

create_doc

Grading

1. field_equals / create_doc

{
  "type": "field_equals",
  "tool": "create_doc",
  "action_id": "create_eligibility_integrity_record",
  "path": "result.ok",
  "value": true,
  "check_id": "riley_129_00"
}

2. field_equals / create_doc

{
  "check_id": "riley_129_01",
  "type": "field_equals",
  "tool": "create_doc",
  "action_id": "create_eligibility_integrity_record",
  "path": "args.title",
  "value": "Expansion Assist second cohort — eligibility integrity decision"
}

3. field_llm_judge / create_doc

{
  "type": "field_llm_judge",
  "tool": "create_doc",
  "action_id": "create_eligibility_integrity_record",
  "path": "args",
  "criterion": "The body is a concise decision record that clearly states that the second Expansion Assist cohort had an operational cap of 18 accounts but was locked at 15 evidence-complete candidates; explains that eligibility was not relaxed and marginal records were not added merely to fill the three-account difference; and frames the unused capacity as the result of preserving eligibility integrity rather than as a delivery or execution failure. Missing or contradicting any of these points must fail.",
  "check_id": "riley_129_02"
}

Complete grading specification

{
  "type": "tool_trace",
  "config": {
    "check_version": 2,
    "today": "2028-01-01",
    "semantic_judge_version": 2,
    "assertions": [
      {
        "type": "field_equals",
        "tool": "create_doc",
        "action_id": "create_eligibility_integrity_record",
        "path": "result.ok",
        "value": true,
        "check_id": "riley_129_00"
      },
      {
        "check_id": "riley_129_01",
        "type": "field_equals",
        "tool": "create_doc",
        "action_id": "create_eligibility_integrity_record",
        "path": "args.title",
        "value": "Expansion Assist second cohort — eligibility integrity decision"
      },
      {
        "type": "field_llm_judge",
        "tool": "create_doc",
        "action_id": "create_eligibility_integrity_record",
        "path": "args",
        "criterion": "The body is a concise decision record that clearly states that the second Expansion Assist cohort had an operational cap of 18 accounts but was locked at 15 evidence-complete candidates; explains that eligibility was not relaxed and marginal records were not added merely to fill the three-account difference; and frames the unused capacity as the result of preserving eligibility integrity rather than as a delivery or execution failure. Missing or contradicting any of these points must fail.",
        "check_id": "riley_129_02"
      }
    ]
  }
}

App state

Download JSON

Source file

tests/riley/129.yaml

SHA-256: 67231e0351e723373cbd1537597c95a6fbe4059432fd446fe6b15954faf9ea76