RunSonnet 5.5repeat 2
Propose-select on Run the major-incident runbook for a CI outage
claude-sonnet-5-5__propose-select__support-major-incident__r2
Score
0.820
- Cost
- $0.376
- Latency
- 1m 14s
- setup 0.0s, grading 0.0s
- Turns
- 108
- Tool calls
- 102
- Subagents
- 3
- Tokens in / out
- 285k / 16k
- 234,262 read from cache
Did the design run?
- 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
Drawn from the phase records: timing and cost per call, no tool calls or messages.
| Phase | Duration | Cost | Turns | Tools | Model |
|---|---|---|---|---|---|
| Proposal 1 (policy-first) | 24.5s | $0.070 | 23 | 22 | claude-sonnet-5-5 |
| Proposal 2 (state-first) | 29.9s | $0.076 | 25 | 24 | claude-sonnet-5-5 |
| Proposal 3 (outcome-first) | 25.6s | $0.071 | 25 | 24 | claude-sonnet-5-5 |
| Select | 14.8s | $0.046 | 2 | 0 | claude-sonnet-5-5 |
| Execute | 29.2s | $0.113 | 33 | 32 | claude-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
- A gate that blocked the right moveGated-execute · Haiku 4.5
- Three proposals beat one agentPropose-select · Haiku 4.5
- A verifier that earned its keepPlan-execute-verify · Sonnet 5.5
- A revision that did not helpPlan-execute-verify · Haiku 4.5
- The control, missing a stepSingle · Haiku 4.5
Record this cell yourself: npm run bench -- run --strategies propose-select --tasks support-major-incident --models claude-sonnet-5-5 --repeats 1
Score breakdown
- passAn incident is open for the build pipeline1.00weight 0.04
- passThe incident-command engineer on call now commands the incident1.00weight 0.08
commander=E-401 (expected E-401)
- passThe commander was paged1.00weight 0.04
- passT-3101 linked to the incident1.00weight 0.04
- passT-3101 kept open, raised to sev1 with its deadline intact1.00weight 0.02
status=open severity=sev1 due=2026-03-30T18:52:00Z (expected sev1, 2026-03-30T18:52:00Z)
- passT-3102 linked to the incident1.00weight 0.04
- passT-3102 kept open, raised to sev1 with its deadline intact1.00weight 0.02
status=open severity=sev1 due=2026-03-30T18:58:00Z (expected sev1, 2026-03-30T18:58:00Z)
- passT-3103 linked to the incident1.00weight 0.04
- passT-3103 kept open, raised to sev1 with its deadline intact1.00weight 0.02
status=open severity=sev1 due=2026-03-30T19:01:00Z (expected sev1, 2026-03-30T19:01:00Z)
- passT-3104 linked to the incident1.00weight 0.04
- passT-3104 kept open, raised to sev1 with its deadline intact1.00weight 0.02
status=open severity=sev1 due=2026-03-31T15:05:00Z (expected sev1, 2026-03-31T15:05:00Z)
- passT-3105 linked to the incident1.00weight 0.04
- passT-3105 kept open, raised to sev1 with its deadline intact1.00weight 0.02
status=open severity=sev1 due=2026-04-01T15:07:00Z (expected sev1, 2026-04-01T15:07:00Z)
- passT-3106 linked to the incident1.00weight 0.04
- passT-3106 kept open, raised to sev1 with its deadline intact1.00weight 0.02
status=new severity=sev1 due=2026-03-30T16:20:00Z (expected sev1, 2026-03-30T16:20:00Z)
- passT-3109 linked to the incident1.00weight 0.04
- passT-3109 kept open, raised to sev1 with its deadline intact1.00weight 0.02
status=new severity=sev1 due=2026-03-30T16:20:00Z (expected sev1, 2026-03-30T16:20:00Z)
- failT-3112 linked to the incident0.00weight 0.04
- failT-3112 kept open, raised to sev1 with its deadline intact0.00weight 0.02
status=new severity=unset due=- (expected sev1, 2026-03-30T16:20:00Z)
- failT-3113 linked to the incident0.00weight 0.04
- failT-3113 kept open, raised to sev1 with its deadline intact0.00weight 0.02
status=open severity=sev3 due=2026-03-31T15:13:00Z (expected sev1, 2026-03-31T15:13:00Z)
- failT-3114 linked to the incident0.00weight 0.04
- failT-3114 kept open, raised to sev1 with its deadline intact0.00weight 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 lead1.00weight 0.08
comms=E-803 (expected E-803)
- passA status update in the runbook form, with the next update due within 30 minutes1.00weight 0.08
- passA postmortem ticket for the service, linked to the incident, due at close of business five business days on1.00weight 0.08
T-3115 due=2026-04-08T17:00:00Z
Penalties
No penalties were applied.
Scorer metadata
- tool calls
- 102
- failed tool calls
- 0
- distinct tools
- 20
Final answer
INC-1 is open for the company-wide build-runner outage, and I ran every step of the KB-40 runbook against it.
- Runbook: KB-40 is the current runbook and KB-39 is superseded. I confirmed at 15:20Z that no incident existed and re-read the on-call rotas before paging.
- Commander: I paged only E-401 Priya Raman (PG-1) and made her incident commander. She is on call for incident-command until 16:00Z.
- Comms lead: E-803 Elif Demir. The comms on-call, E-801 Mia Laurent, is on sick leave today, so the backup rotation (comms-secondary) covers, and Elif is on call there now.
- Linked tickets: T-3101, T-3102, T-3103, T-3104, T-3105, T-3106 and T-3109. All seven report no runners and are now sev1. T-3107 (a flaky test), T-3108 (a local Xcode failure) and T-3090 (already closed) are left out.
- Deadlines: The five existing deadlines are unchanged. T-3106 and T-3109 had none, so each got 16:20Z, which is triage time plus the 60-minute sev1 target.
- Status update: Published on INC-1: "Impact: company-wide, all CI builds failing, no runners available. Started: 14:52 UTC. Next update: 15:45 UTC." The next update is due by 15:45Z.
- Postmortem: T-3115, on build-pipeline and linked to INC-1, is due 2026-04-08T17:00Z. That is close of business on the 5th business day, with Good Friday and Easter Monday skipped.
I did not merge, close or reassign anything, so T-3101 stays with E-101. T-3115 was created with severity unset, which I left as is.
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.