DolphinBench

Test 016

Jan 1, 2028 / 1 fact

YAML

Request

The January sanitized campaign snapshot is due Tuesday, January 4, 2028. Create a concise source-readiness checklist covering the check timing, required source details, and late or missing-certification handling, and DM Owen the same guidance as a January-cycle reminder.

Required memory

Fact 241

The January 15, 2027 decision requires Owen to run a monthly source-readiness check at noon Central on the business day before each sanitized campaign snapshot is due. Every required source must show a named owner, exact certified through-date, and certification ETA; Owen must mark the cycle at risk if an ETA is later than the snapshot deadline and open the snapshot visibly incomplete rather than use stale data if certification remains absent on the due date.

Source evidence (1)

003931Jan 15, 2027 / 09:10 UTC-06:00

We’ve made the January 15 decision. Owen will run a monthly source-readiness check at noon Central on the business day before each sanitized campaign snapshot is due. Each required source must show a named owner, exact certified through-date, and certification ETA. If an ETA falls after the snapshot deadline, Owen marks the cycle at risk; if certification is still absent on the due date, he opens the snapshot visibly incomplete rather than using stale data. This does not change the second-business-day deadline, owner-certification requirement, freshness limits, source meanings, existing authority, or fail-closed handling. Update `Q1 2027 Lifecycle Systems operating priorities — staff-review draft` with the final decision and post the concise implementation summary in `#lifecycle-systems`.

Message 003931 in history

Expected tool calls

  • create_doc
  • send_slack_dm

Grading

1. field_equals / create_doc
{
  "type": "field_equals",
  "tool": "create_doc",
  "action_id": "checklist_doc",
  "path": "result.ok",
  "value": true,
  "check_id": "riley_016_00"
}
2. field_equals / send_slack_dm
{
  "type": "field_equals",
  "tool": "send_slack_dm",
  "action_id": "january_reminder",
  "path": "result.ok",
  "value": true,
  "check_id": "riley_016_01"
}
3. field_equals / send_slack_dm
{
  "type": "field_equals",
  "tool": "send_slack_dm",
  "action_id": "january_reminder",
  "path": "args.user",
  "value": "owen",
  "check_id": "riley_016_02"
}
4. field_llm_judge / create_doc
{
  "type": "field_llm_judge",
  "tool": "create_doc",
  "action_id": "checklist_doc",
  "path": "args",
  "criterion": "The body is a concise, usable source-readiness checklist for the January 4, 2028 snapshot. It identifies the check time as noon Central on the business day before that deadline (Monday, January 3, 2028; either relational or explicit-date wording is acceptable), requires every required source to show a named owner, an exact certified-through date, and a certification ETA, says an ETA after the snapshot deadline makes the cycle at risk, and says certification still absent on the due date requires opening the snapshot in a visibly incomplete state rather than using stale data.",
  "check_id": "riley_016_03"
}
5. field_llm_judge / send_slack_dm
{
  "type": "field_llm_judge",
  "tool": "send_slack_dm",
  "action_id": "january_reminder",
  "path": "args",
  "criterion": "The message is a January-cycle readiness reminder for the January 4, 2028 snapshot. It identifies the check time as noon Central on the business day before that deadline (Monday, January 3, 2028; either relational or explicit-date wording is acceptable), requires every required source to show a named owner, an exact certified-through date, and a certification ETA, says an ETA after the snapshot deadline makes the cycle at risk, and says certification still absent on the due date requires opening the snapshot in a visibly incomplete state rather than using stale data.",
  "check_id": "riley_016_04"
}
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": "checklist_doc",
        "path": "result.ok",
        "value": true,
        "check_id": "riley_016_00"
      },
      {
        "type": "field_equals",
        "tool": "send_slack_dm",
        "action_id": "january_reminder",
        "path": "result.ok",
        "value": true,
        "check_id": "riley_016_01"
      },
      {
        "type": "field_equals",
        "tool": "send_slack_dm",
        "action_id": "january_reminder",
        "path": "args.user",
        "value": "owen",
        "check_id": "riley_016_02"
      },
      {
        "type": "field_llm_judge",
        "tool": "create_doc",
        "action_id": "checklist_doc",
        "path": "args",
        "criterion": "The body is a concise, usable source-readiness checklist for the January 4, 2028 snapshot. It identifies the check time as noon Central on the business day before that deadline (Monday, January 3, 2028; either relational or explicit-date wording is acceptable), requires every required source to show a named owner, an exact certified-through date, and a certification ETA, says an ETA after the snapshot deadline makes the cycle at risk, and says certification still absent on the due date requires opening the snapshot in a visibly incomplete state rather than using stale data.",
        "check_id": "riley_016_03"
      },
      {
        "type": "field_llm_judge",
        "tool": "send_slack_dm",
        "action_id": "january_reminder",
        "path": "args",
        "criterion": "The message is a January-cycle readiness reminder for the January 4, 2028 snapshot. It identifies the check time as noon Central on the business day before that deadline (Monday, January 3, 2028; either relational or explicit-date wording is acceptable), requires every required source to show a named owner, an exact certified-through date, and a certification ETA, says an ETA after the snapshot deadline makes the cycle at risk, and says certification still absent on the due date requires opening the snapshot in a visibly incomplete state rather than using stale data.",
        "check_id": "riley_016_04"
      }
    ]
  }
}
App stateDownload JSON
Source file

tests/riley/016.yaml

SHA-256: d1a709abe9ecb7a97e25ef090c18ec9a279b43139421d4463cb8de4093c6d2e6