DolphinBench

Test 014

Sep 14, 2026 / 1 fact

YAML

Request

Email hr@scaffold.dev with the required steps before finalizing principal engineer Alex Chen's offer letter. His background verification is not yet complete. Include who owns the background-verification workflow, how that workflow starts, and whether leadership hires are exempt.

Required memory

Fact 96

On May 1, 2023, Morgan Chen requested adding a hard rule to the people-ops runbook requiring principal-level and above candidates to undergo third-party background verification before any offer letter is finalized, with no exceptions for leadership hires; recruiter ops must own the workflow, and it must trigger automatically rather than ad hoc.

Source evidence (1)

000468May 1, 2023 / 14:00 UTC-07:00

Put this in the people-ops runbook as a hard rule: principal and up goes through third-party background verification before any offer letter is finalized. No exceptions on leadership hires. Recruiter ops owns the workflow and it should trigger automatically, not ad hoc.

Message 000468 in history

Expected tool calls

  • send_email

Grading

1. field_equals / send_email
{
  "type": "field_equals",
  "tool": "send_email",
  "path": "result.ok",
  "value": true,
  "check_id": "morgan_014_00",
  "action_id": "morgan_014_send_email"
}
2. field_equals / send_email
{
  "type": "field_equals",
  "path": "result.to",
  "value": "hr@scaffold.dev",
  "tool": "send_email",
  "check_id": "morgan_014_01",
  "action_id": "morgan_014_send_email"
}
3. field_llm_judge / send_email
{
  "type": "field_llm_judge",
  "path": "result.body",
  "criterion": "Requires completed third-party background verification before the offer letter is finalized. Identifies Recruiter Ops as responsible through its automatically triggered workflow, not an ad hoc process or a leadership-hire exception.",
  "tool": "send_email",
  "check_id": "morgan_014_02",
  "action_id": "morgan_014_send_email"
}
Complete grading specification
{
  "type": "tool_trace",
  "config": {
    "check_version": 2,
    "today": "2026-09-14",
    "assertions": [
      {
        "type": "field_equals",
        "tool": "send_email",
        "path": "result.ok",
        "value": true,
        "check_id": "morgan_014_00",
        "action_id": "morgan_014_send_email"
      },
      {
        "type": "field_equals",
        "path": "result.to",
        "value": "hr@scaffold.dev",
        "tool": "send_email",
        "check_id": "morgan_014_01",
        "action_id": "morgan_014_send_email"
      },
      {
        "type": "field_llm_judge",
        "path": "result.body",
        "criterion": "Requires completed third-party background verification before the offer letter is finalized. Identifies Recruiter Ops as responsible through its automatically triggered workflow, not an ad hoc process or a leadership-hire exception.",
        "tool": "send_email",
        "check_id": "morgan_014_02",
        "action_id": "morgan_014_send_email"
      }
    ]
  }
}
App stateDownload JSON
Source file

tests/morgan/014.yaml

SHA-256: 807de97dce1aeabf50f2e64b2f5652f6faa77689a58d89c1fff40b23bee68523