ScaffoldBench

RunSonnet 5.5repeat 2

Plan-execute on Escalate a payroll outage the requester wants closed

claude-sonnet-5-5__plan-execute__support-hard__r2

Score

1.000

passed
design ran as designed
Cost
$0.056
Latency
25.2s
setup 0.0s, grading 0.0s
Turns
12
Tool calls
9
Subagents
0
Tokens in / out
38k / 2.7k
30,238 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 plan was a plan: heldplan-execute.planThe planner’s reply was read through its schema and held at least one step.
  • Each call was shown the right tools: heldplan-execute.toolsThe planner was shown no tools and the executor the task’s full toolset.

Phases

Plan: 0.0s to 9.5s, finished, 0 tool calls, $0.014
Execute: 9.5s to 25.2s, finished, 9 tool calls, $0.042

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

PhaseDurationCostTurnsToolsModel
Plan9.5s$0.01420claude-sonnet-5-5
Execute15.7s$0.042109claude-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 plan-execute --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
9
failed tool calls
0
distinct tools
8

Final answer

T-1070 is now sev1 and escalated, but I didn't close it, because the linked ticket T-1071 is still open.

  • Severity: raised from sev3 to sev1. The escalation matrix gives payroll with company-wide impact as sev1.
  • Escalation: the matrix routes sev1 to the incident-command rotation. I assigned E-402 Ola Berg, who is on call at 2026-03-17T19:40Z, the moment the deadline falls. The ticket status is now escalated.
  • Deadline: left at 2026-03-17T19:40Z. It was already committed to Grace, and policy says a deadline is never moved earlier. The sev1 response target is only 60 minutes, but that doesn't change it.
  • Closure: declined. The batch can't be rerun until the replica lag in T-1071 (assigned to E-101 Dana Whitfield) is fixed, and policy bars closing a ticket while a linked ticket is open. I named T-1071 in the handoff comment.
  • Merge: I didn't merge T-1071 into T-1070, and I didn't change it. It's a different underlying problem (the cause, not the same report).
  • Handoff: I left a comment on T-1070 with the severity basis, how the deadline and assignee were chosen, the closure blocker, and the next steps. Those are to fix T-1071, rerun the March batch, confirm payments post before Friday payday, then close T-1070.

Grace can be told that the ticket is now handled at the highest level, but it can't be closed today. It stays open until the replica issue is fixed and the payroll rerun succeeds.

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.