ScaffoldBench

RunOpus 5.5repeat 2

Propose-select on Run the major-incident runbook for a CI outage

claude-opus-5-5__propose-select__support-major-incident__r2

Score

0.820

wrong answer
design ran as designed
Cost
$0.916
Latency
2m 2s
setup 0.0s, grading 0.0s
Turns
131
Tool calls
125
Subagents
3
Tokens in / out
369k / 22k
305,911 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.
  • All three proposers ran: heldpropose-select.proposalsThree proposers ran in parallel, each from its own angle, and none was starved while another ran.
  • The selector chose: heldpropose-select.selectionThe selector picked a proposal by a valid number, read from its schema reply and not a default.
  • Each call was shown the right tools: heldpropose-select.toolsEach proposer was shown the read-only tools, the selector none, and the executor the full toolset.

Phases

Proposal 1 (policy-first): 0.0s to 41.8s, finished, 28 tool calls, $0.173
Proposal 2 (state-first): 0.0s to 50.9s, finished, 31 tool calls, $0.206
Proposal 3 (outcome-first): 0.0s to 49.4s, finished, 29 tool calls, $0.205
Select: 50.9s to 1m 17s, finished, 0 tool calls, $0.103
Execute: 1m 17s to 2m 2s, finished, 37 tool calls, $0.228

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

PhaseDurationCostTurnsToolsModel
Proposal 1 (policy-first)41.8s$0.1732928claude-opus-5-5
Proposal 2 (state-first)50.9s$0.2063231claude-opus-5-5
Proposal 3 (outcome-first)49.4s$0.2053029claude-opus-5-5
Select25.8s$0.10320claude-opus-5-5
Execute45.5s$0.2283837claude-opus-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 propose-select --tasks support-major-incident --models claude-opus-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=open 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=open 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
125
failed tool calls
1
distinct tools
21

Final answer

I've opened the build outage as major incident INC-1 and run it under the current runbook, KB-40. The clock was 2026-03-30 15:20 UTC. I didn't use KB-39: it has been replaced, and it would have paged the wrong person and pulled committed deadlines earlier. Before acting I rechecked the current state, and it matched what the plan assumed.

  1. Incident opened: INC-1 was created at 15:20, before anyone was contacted. The matrix rates a company-wide build-pipeline outage as sev1, and no knowledge-base article changes that.
  2. Incident commander: I sent one page, to E-401 Priya Raman, who is on call for incident command now and not on leave. She is the commander. There were no other pages.
  3. Reports linked: I linked the seven reports of jobs never getting a runner: T-3101, T-3102, T-3103, T-3104, T-3105, T-3106 and T-3109. I left out T-3107 (the known flaky payments test), T-3108 (a local Xcode build failure) and T-3090 (already closed). I didn't merge any duplicates, because the runbook says to link every report.

  4. Severity and deadlines: All seven reports are now sev1.

    • The five existing deadlines are unchanged: 18:52, 18:58, 19:01, 03-31 15:05 and 04-01 15:07.
    • T-3106 and T-3109 had no deadline. Each now has 15:20 plus the 60-minute sev1 target, which is 16:20.
    • Both are assigned to E-101 Dana Whitfield, who is on call for platform-primary (the team that owns builds) at 16:20 and not on leave.
  5. Communications lead: The comms rota names E-801 Mia Laurent, but she is on sick leave. I named E-803 Elif Demir, who is on call for the backup comms rotation.
  6. First status update (published 15:20): "Impact: All CI builds company-wide; no build runners are available in any pool, so all teams' builds and releases are blocked. Started: 14:52 UTC. Next update: 15:50 UTC." 14:52 is when the earliest report, T-3101, came in.
  7. Postmortem ticket: T-3115 is linked to INC-1 and due 2026-04-08 at 17:00 UTC. That is close of business on the 5th business day after the incident opened, skipping the weekend, Good Friday (04-03) and Easter Monday (04-06).

I closed nothing and moved no deadline earlier. The postmortem ticket has no severity set, because the runbook doesn't call for one. The next status update is due at 15:50 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.