DolphinBench

Test 003

Jan 1, 2028 / 2 facts

YAML

Request

Create a current registration record for Kibo. Include the issuing authority's effective dates, the matching Cherrywood property-file status, and whether any duplicate packet or additional document is still required.

Required memory

Fact 259

The registration authority finished processing Kibo’s timely June 1, 2027 packet and issued his renewed registration effective June 16, 2027 through June 15, 2028; no duplicate packet or additional document is required, and the interim-proof matter and authority review are resolved.

Source evidence (1)

004379Jun 30, 2027 / 11:05 UTC-05:00

The registration authority finished processing Kibo’s timely June 1 packet and issued his renewed registration effective June 16, 2027 through June 15, 2028. No duplicate packet or additional document is required. I uploaded the issued registration to Cherrywood, and the resident portal now marks his property-file registration current through June 15, 2028. The interim-proof matter and authority review are resolved. His rabies certificate remains current through October 31, 2027.

Message 004379 in history

Fact 260

Cherrywood’s resident portal now marks Kibo’s property-file registration current through June 15, 2028.

Source evidence (1)

004379Jun 30, 2027 / 11:05 UTC-05:00

The registration authority finished processing Kibo’s timely June 1 packet and issued his renewed registration effective June 16, 2027 through June 15, 2028. No duplicate packet or additional document is required. I uploaded the issued registration to Cherrywood, and the resident portal now marks his property-file registration current through June 15, 2028. The interim-proof matter and authority review are resolved. His rabies certificate remains current through October 31, 2027.

Message 004379 in history

Expected tool calls

  • create_doc

Grading

1. field_equals / create_doc
{
  "type": "field_equals",
  "tool": "create_doc",
  "action_id": "create_registration_record",
  "path": "result.ok",
  "value": true,
  "check_id": "riley_003_00"
}
2. field_llm_judge / create_doc
{
  "type": "field_llm_judge",
  "tool": "create_doc",
  "action_id": "create_registration_record",
  "path": "args",
  "criterion": "The record says Kibo's registration was issued for June 16, 2027 through June 15, 2028, no duplicate packet or additional document is required, and Cherrywood marks the property-file registration current through June 15, 2028.",
  "check_id": "riley_003_01"
}
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_registration_record",
        "path": "result.ok",
        "value": true,
        "check_id": "riley_003_00"
      },
      {
        "type": "field_llm_judge",
        "tool": "create_doc",
        "action_id": "create_registration_record",
        "path": "args",
        "criterion": "The record says Kibo's registration was issued for June 16, 2027 through June 15, 2028, no duplicate packet or additional document is required, and Cherrywood marks the property-file registration current through June 15, 2028.",
        "check_id": "riley_003_01"
      }
    ]
  }
}
App stateDownload JSON
Source file

tests/riley/003.yaml

SHA-256: b066f471e22bd7b9a4fe5af8941886e8d2d2a98ef28361b2efc91c5bb5886438