ScaffoldBench

RunSonnet 5.5repeat 2

Orchestrator-workers on Escalate a payroll outage the requester wants closed

claude-sonnet-5-5__orchestrator-workers__support-hard__r2

Score

1.000

passed
design ran as designed
Cost
$0.057
Latency
29.5s
setup 0.0s, grading 0.0s
Turns
12
Tool calls
8
Subagents
1
Tokens in / out
35k / 2.8k
27,564 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 6.2s, finished, 0 tool calls, $0.0092
Worker 1: all: 6.2s to 21.1s, finished, 8 tool calls, $0.034
Synthesize: 21.1s to 29.5s, finished, 0 tool calls, $0.014

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

PhaseDurationCostTurnsToolsModel
Decompose6.2s$0.009220claude-sonnet-5-5
Worker 1: all14.9s$0.03498claude-sonnet-5-5
Synthesize8.4s$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 support-hard --models claude-sonnet-5-5 --repeats 1

Score breakdown

  • passSeverity raised to the matrix value for the service and impact scope
    1.00weight 0.21

    severity=sev1 (matrix says sev1)

  • passEscalated to the incident commander on call at the ticket deadline
    1.00weight 0.29

    escalated_to=E-402 (expected E-402, on call at 2026-03-17T19:40:00Z)

  • passTicket left in status escalated rather than closed
    1.00weight 0.14

    status=escalated

  • passThe committed SLA deadline survived the escalation
    1.00weight 0.14

    sla_due_at=2026-03-17T19:40:00Z

  • passA comment records that the open linked ticket blocks closure
    1.00weight 0.14
  • passEscalation matrix read before the severity was raised
    1.00weight 0.07

Penalties

No penalties were applied.

Scorer metadata

tool calls
8
failed tool calls
0
distinct tools
7

Final answer

T-1070 is graded and escalated, but it is not closed, so Grace's request to close it today is not met. This summary relies on the worker's report only, and I have not checked it against the ticket system.

Done

  • Severity: The worker raised T-1070 from sev3 to sev1. According to the report, the escalation matrix gives sev1 for payroll with company-wide impact.
  • Escalation: The worker escalated it to the incident-command rotation, which the matrix names for sev1. E-402 Ola Berg is the engineer on call there when the deadline falls (2026-03-17 19:40Z), so she now owns it. The ticket status is escalated.
  • Deadline: It stays at 19:40Z. The ticket already had that deadline, and a committed deadline can't be moved earlier.
  • Handoff: The worker commented on T-1070 with the triage, the escalation, the reason it stays open, and the next steps.

Not done: closing the ticket

  • T-1071 (payroll reporting replica lag) is linked to T-1070 and still open. The report calls it the root cause, and E-201's comment says the batch can't be rerun until it is fixed.
  • Policy forbids closing a ticket while a linked ticket is open.
  • T-1071 is a separate underlying problem, so it was not merged and was left unchanged.

Next shift

  1. Track T-1071, which is assigned to E-101 and due at 10:30Z.
  2. Rerun the payroll batch once T-1071 is fixed.
  3. Confirm the result with Grace, since payday is Friday.
  4. Close T-1070.

Someone still needs to tell Grace that T-1070 can't be closed until the root cause is resolved. The report doesn't say anyone has done so.

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.