ScaffoldBench

RunSonnet 5.5repeat 2

Orchestrator-workers on Add checked bags and collect the right fee

claude-sonnet-5-5__orchestrator-workers__airline-easy__r2

Score

1.000

passed
design ran as designed
Cost
$0.053
Latency
24.5s
setup 0.0s, grading 0.0s
Turns
11
Tool calls
7
Subagents
1
Tokens in / out
41k / 2.1k
31,621 read from cache

Did the design run?

every check held
  • Run completed: heldharness.run-completedThe environment was built, the architecture returned and grading finished, none of them throwing.
  • No API failures: heldharness.apiNo model call failed at the API.
  • Every call ended: heldharness.calls-endedEvery call ended with a result, or was stopped by the run's own limits.
  • No harness tool denied: heldharness.harness-toolsThe sandbox guard never denied a harness tool, nor a tool the session itself offered.
  • Tool servers connected: heldharness.mcp-connectedEvery tool server the harness gave a call was connected when the session started.
  • No foreign tool servers: heldharness.mcp-hermeticNo call loaded a tool server the harness did not give it.
  • Builtin tools: heldharness.builtin-toolsA check recorded with the run.
  • Structured replies arrived: heldharness.structured-outputEvery call that asked for structured output and ran to the end got it.
  • No silent fallbacks: heldharness.not-degradedNo phase fell back on a default (an empty plan, a default winner) instead of its output.
  • The task was decomposed: heldorchestrator-workers.decompositionThe orchestrator’s reply was read through its schema and held at least one subtask.
  • The fan-out happened: heldorchestrator-workers.fan-outEvery subtask got a worker, every worker ran, and the recorded subagent count matches.
  • The synthesis wrote the answer: heldorchestrator-workers.synthesisThe final answer came from the synthesis call, not from a worker’s report standing in for it.
  • Each call was shown the right tools: heldorchestrator-workers.toolsThe orchestrator and the synthesis were shown no tools and every worker the task’s full toolset.

Phases

Decompose: 0.0s to 5.4s, finished, 0 tool calls, $0.011
Worker 1: all: 5.4s to 17.7s, finished, 7 tool calls, $0.028
Synthesize: 17.7s to 24.5s, finished, 0 tool calls, $0.014

Drawn from the phase records: timing and cost per call, no tool calls or messages.

PhaseDurationCostTurnsToolsModel
Decompose5.4s$0.01120claude-sonnet-5-5
Worker 1: all12.4s$0.02887claude-sonnet-5-5
Synthesize6.8s$0.01410claude-sonnet-5-5

The call-by-call replay is not published for this run. Traces are large, so the repository keeps 9 of this run set’s 900 and leaves the rest out; the score, phases and final answer on this page come from the committed run record.

Runs you can replay

Record this cell yourself: npm run bench -- run --strategies orchestrator-workers --tasks airline-easy --models claude-sonnet-5-5 --repeats 1

Score breakdown

  • passReservation holds three checked bags and is otherwise untouched
    1.00weight 0.43

    checked_bags=3, status=confirmed

  • passExactly one baggage fee, for the bags beyond the free allowance
    1.00weight 0.57

    expected 4500 cents for 1 chargeable bag(s), charged 4500

Penalties

No penalties were applied.

Scorer metadata

tool calls
7
failed tool calls
0
distinct tools
7

Final answer

I added two checked bags to R1042 and charged Rosa Mendel $45.00 in a single baggage_fee ledger entry (L1). This comes from the worker's report only. I have no tools here and could not check it myself.

  • Bags: The reservation went from 1 to 3 checked bags, which is the policy maximum.
  • Why $45.00: The worker said her economy fare includes 1 free bag and her silver tier adds 1 more, so 2 bags are free. Only the third bag is chargeable, and it is the first chargeable bag, at $45.00. The worker did not charge for the other new bag.
  • Before charging: The ledger had no prior entries, and flight AS420 SFO→SEA showed as scheduled.

Flight date: AS420 is dated 2026-08-11, but today is 2026-10-06, so the flight date is in the past. The system still shows the flight as scheduled and the reservation as confirmed, so the worker went ahead. Someone should confirm the date is correct, since it may mean a data error or that the charge should not have been taken.

The report does not say which tool calls the worker made to confirm the allowance and prices, only the figures themselves.

Provenance of the numbers on this page

run set
scaffold-or-scale-v1
models
Haiku 4.5, Sonnet 5.5, Opus 5.5
effort
high
repeats
2
fixtures
replay
runs
900 / 900
strategy cost
$354.44
judge cost
$9.19
wall clock
9h 48m
git
fce5807b2d

Cost excludes judge calls, which are tracked separately so a scoring model never flatters or penalises an architecture. Every figure on this site is read from results/ at build time.