ScaffoldBench

RunSonnet 5.5repeat 2

Gated-execute on Run the major-incident runbook for a CI outage

claude-sonnet-5-5__gated-execute__support-major-incident__r2

Score

0.820

wrong answer
design ran as designed
Cost
$0.534
Latency
2m 57s
setup 0.0s, grading 0.0s
Turns
159
Tool calls
123
Subagents
17
Tokens in / out
499k / 16k
405,830 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.
  • Every state change was reviewed: heldgated-execute.reviewedEvery call that changed the world had been allowed by a review before it ran, counted from the world’s own journal.
  • Every decision was read: heldgated-execute.decisionsEvery review that ran to its end gave its decision through the schema; an unreadable one is a denial.
  • Each call was shown the right tools: heldgated-execute.toolsThe executor was shown the task’s full toolset and each reviewer the read-only tools.

Phases

Execute: 0.0s to 2m 57s, finished, 39 tool calls, $0.116
Review 1: create_incident: 11.7s to 18.7s, finished, 3 tool calls, $0.019
Review 2: page_engineer: 20.9s to 27.3s, finished, 6 tool calls, $0.020
Review 3: assign_incident_role: 27.3s to 34.2s, finished, 5 tool calls, $0.019
Review 4: link_to_incident: 34.2s to 45.7s, finished, 12 tool calls, $0.041
Review 5: assign_incident_role: 45.8s to 57.6s, finished, 11 tool calls, $0.035
Review 6: set_severity: 59.7s to 1m 6s, finished, 3 tool calls, $0.020
Review 7: set_severity: 1m 6s to 1m 12s, finished, 3 tool calls, $0.019
Review 8: set_severity: 1m 12s to 1m 20s, finished, 3 tool calls, $0.020
Review 9: set_severity: 1m 20s to 1m 28s, finished, 3 tool calls, $0.022
Review 10: set_severity: 1m 28s to 1m 36s, finished, 2 tool calls, $0.019
Review 11: set_severity: 1m 37s to 1m 47s, finished, 4 tool calls, $0.023
Review 12: set_severity: 1m 47s to 1m 55s, finished, 5 tool calls, $0.027
Review 13: assign_incident_role: 1m 57s to 2m 7s, finished, 7 tool calls, $0.029
Review 14: post_status_update: 2m 10s to 2m 19s, finished, 4 tool calls, $0.025
Review 15: create_ticket: 2m 19s to 2m 30s, finished, 4 tool calls, $0.025
Review 16: extend_sla: 2m 34s to 2m 41s, finished, 4 tool calls, $0.024
Review 17: extend_sla: 2m 41s to 2m 52s, finished, 6 tool calls, $0.032

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

PhaseDurationCostTurnsToolsModel
Execute2m 57s$0.1164039claude-sonnet-5-5
Review 1: create_incident7.0s$0.01953claude-sonnet-5-5
Review 2: page_engineer6.4s$0.02086claude-sonnet-5-5
Review 3: assign_incident_role6.9s$0.01975claude-sonnet-5-5
Review 4: link_to_incident11.5s$0.0411412claude-sonnet-5-5
Review 5: assign_incident_role11.8s$0.0351311claude-sonnet-5-5
Review 6: set_severity6.6s$0.02053claude-sonnet-5-5
Review 7: set_severity6.1s$0.01953claude-sonnet-5-5
Review 8: set_severity7.5s$0.02053claude-sonnet-5-5
Review 9: set_severity7.6s$0.02253claude-sonnet-5-5
Review 10: set_severity8.4s$0.01942claude-sonnet-5-5
Review 11: set_severity9.9s$0.02364claude-sonnet-5-5
Review 12: set_severity8.5s$0.02775claude-sonnet-5-5
Review 13: assign_incident_role9.7s$0.02997claude-sonnet-5-5
Review 14: post_status_update8.9s$0.02564claude-sonnet-5-5
Review 15: create_ticket10.9s$0.02564claude-sonnet-5-5
Review 16: extend_sla7.4s$0.02464claude-sonnet-5-5
Review 17: extend_sla10.8s$0.03286claude-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 gated-execute --tasks support-major-incident --models claude-sonnet-5-5 --repeats 1

Score breakdown

  • passAn incident is open for the build pipeline
    1.00weight 0.04
  • passThe incident-command engineer on call now commands the incident
    1.00weight 0.08

    commander=E-401 (expected E-401)

  • passThe commander was paged
    1.00weight 0.04
  • passT-3101 linked to the incident
    1.00weight 0.04
  • passT-3101 kept open, raised to sev1 with its deadline intact
    1.00weight 0.02

    status=open severity=sev1 due=2026-03-30T18:52:00Z (expected sev1, 2026-03-30T18:52:00Z)

  • passT-3102 linked to the incident
    1.00weight 0.04
  • passT-3102 kept open, raised to sev1 with its deadline intact
    1.00weight 0.02

    status=open severity=sev1 due=2026-03-30T18:58:00Z (expected sev1, 2026-03-30T18:58:00Z)

  • passT-3103 linked to the incident
    1.00weight 0.04
  • passT-3103 kept open, raised to sev1 with its deadline intact
    1.00weight 0.02

    status=open severity=sev1 due=2026-03-30T19:01:00Z (expected sev1, 2026-03-30T19:01:00Z)

  • passT-3104 linked to the incident
    1.00weight 0.04
  • passT-3104 kept open, raised to sev1 with its deadline intact
    1.00weight 0.02

    status=open severity=sev1 due=2026-03-31T15:05:00Z (expected sev1, 2026-03-31T15:05:00Z)

  • passT-3105 linked to the incident
    1.00weight 0.04
  • passT-3105 kept open, raised to sev1 with its deadline intact
    1.00weight 0.02

    status=open severity=sev1 due=2026-04-01T15:07:00Z (expected sev1, 2026-04-01T15:07:00Z)

  • passT-3106 linked to the incident
    1.00weight 0.04
  • passT-3106 kept open, raised to sev1 with its deadline intact
    1.00weight 0.02

    status=new severity=sev1 due=2026-03-30T16:20:00Z (expected sev1, 2026-03-30T16:20:00Z)

  • passT-3109 linked to the incident
    1.00weight 0.04
  • passT-3109 kept open, raised to sev1 with its deadline intact
    1.00weight 0.02

    status=new severity=sev1 due=2026-03-30T16:20:00Z (expected sev1, 2026-03-30T16:20:00Z)

  • failT-3112 linked to the incident
    0.00weight 0.04
  • failT-3112 kept open, raised to sev1 with its deadline intact
    0.00weight 0.02

    status=new severity=unset due=- (expected sev1, 2026-03-30T16:20:00Z)

  • failT-3113 linked to the incident
    0.00weight 0.04
  • failT-3113 kept open, raised to sev1 with its deadline intact
    0.00weight 0.02

    status=open severity=sev3 due=2026-03-31T15:13:00Z (expected sev1, 2026-03-31T15:13:00Z)

  • failT-3114 linked to the incident
    0.00weight 0.04
  • failT-3114 kept open, raised to sev1 with its deadline intact
    0.00weight 0.02

    status=open severity=sev3 due=2026-03-31T15:16:00Z (expected sev1, 2026-03-31T15:16:00Z)

  • passThe engineer covering comms right now is the communications lead
    1.00weight 0.08

    comms=E-803 (expected E-803)

  • passA status update in the runbook form, with the next update due within 30 minutes
    1.00weight 0.08
  • passA postmortem ticket for the service, linked to the incident, due at close of business five business days on
    1.00weight 0.08

    T-3115 due=2026-04-08T17:00:00Z

Penalties

No penalties were applied.

Scorer metadata

tool calls
123
failed tool calls
2
distinct tools
19

Final answer

I opened INC-1 for the company-wide build outage and ran the current major-incident runbook (KB-40) in order. The older runbook KB-39 is marked superseded, so I ignored it.

  • Incident: INC-1 is open on build-pipeline.
  • Commander: I paged E-401 Priya Raman, the only person paged. She is the incident-command engineer on call right now, and she is now the commander.
  • Linked reports: T-3101, T-3102, T-3103, T-3104, T-3105, T-3106 and T-3109 are all reports of the outage and are linked to INC-1.
  • Left out: T-3107 is a flaky test and T-3108 is a local Xcode problem, so neither is part of the outage and I didn't touch them.
  • Severity: I raised all seven linked reports to sev1. The matrix gives sev1 for company-wide impact on build-pipeline. Existing deadlines are unchanged.
  • Missing deadlines: T-3106 and T-3109 had none, so I set them to 16:20 UTC. That is triage time plus the 60-minute sev1 target.
  • Comms lead: E-803 Elif Demir. The comms rotation's on-call, E-801 Mia Laurent, is on sick leave today, so the backup rotation (comms-secondary) covers, and E-803 is on call there now. The reviewer first blocked this step because it comes after the severity step in the runbook, so I raised the severities first and then retried.
  • Status update: I posted "Impact: All teams company-wide… Started: 14:52 UTC. Next update: 15:45 UTC." The start time is when the earliest report came in.
  • Postmortem: T-3115 is linked to INC-1 and due 2026-04-08 at 17:00 UTC. That is the 5th business day after today, skipping the Good Friday (Apr 3) and Easter Monday (Apr 6) holidays.

The next status update is due by 15:45 UTC.

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.