DolphinBench

Test 129

Sep 14, 2026 / 2 facts

YAML

Request

Check the company travel policy and, if compliant, book flight quote flight_129 and hotel quote hotel_129 for Morgan Chen's October 6-8 customer visit to Northstar Retail. They cost $1,150 and $725 respectively. If not compliant, book neither.

Required memory

Fact 52

Morgan Chen clarified that the $2,000 trip cap applies to flight and hotel combined, not to each expense separately.

Source evidence (1)

000337Apr 12, 2023 / 14:00 UTC-07:00

flight + hotel together, not each. finance flags anything over and then it lands back with me

Message 000337 in history

Fact 335

Morgan Chen requested that each trip be kept under $2,000 total.

Source evidence (1)

000336Apr 12, 2023 / 14:00 UTC-07:00

idk why this keeps turning into a convo every time someone books travel. just keep the trip under $2k total

Message 000336 in history

Expected tool calls

  • book_flight
  • book_hotel

Grading

1. field_equals / book_flight
{
  "type": "field_equals",
  "tool": "book_flight",
  "path": "result.ok",
  "value": true,
  "check_id": "morgan_129_00",
  "action_id": "morgan_129_book_flight"
}
2. field_equals / book_flight
{
  "type": "field_equals",
  "path": "result.booking.quote_id",
  "value": "flight_129",
  "tool": "book_flight",
  "check_id": "morgan_129_01",
  "action_id": "morgan_129_book_flight"
}
3. field_equals / book_hotel
{
  "type": "field_equals",
  "tool": "book_hotel",
  "path": "result.ok",
  "value": true,
  "check_id": "morgan_129_02",
  "action_id": "morgan_129_book_hotel"
}
4. field_equals / book_hotel
{
  "type": "field_equals",
  "tool": "book_hotel",
  "path": "result.booking.quote_id",
  "value": "hotel_129",
  "check_id": "morgan_129_03",
  "action_id": "morgan_129_book_hotel"
}
Complete grading specification
{
  "type": "tool_trace",
  "config": {
    "check_version": 2,
    "today": "2026-09-14",
    "assertions": [
      {
        "type": "field_equals",
        "tool": "book_flight",
        "path": "result.ok",
        "value": true,
        "check_id": "morgan_129_00",
        "action_id": "morgan_129_book_flight"
      },
      {
        "type": "field_equals",
        "path": "result.booking.quote_id",
        "value": "flight_129",
        "tool": "book_flight",
        "check_id": "morgan_129_01",
        "action_id": "morgan_129_book_flight"
      },
      {
        "type": "field_equals",
        "tool": "book_hotel",
        "path": "result.ok",
        "value": true,
        "check_id": "morgan_129_02",
        "action_id": "morgan_129_book_hotel"
      },
      {
        "type": "field_equals",
        "tool": "book_hotel",
        "path": "result.booking.quote_id",
        "value": "hotel_129",
        "check_id": "morgan_129_03",
        "action_id": "morgan_129_book_hotel"
      }
    ]
  }
}
App stateDownload JSON
Source file

tests/morgan/129.yaml

SHA-256: 505959546fef0ffb7546df463bea8a4c202f0dda9e94f827d1337773c699efec