ScaffoldBench

RunSonnet 5.5repeat 2

Propose-select on Consolidate duplicate CRM reports

claude-sonnet-5-5__propose-select__support-medium__r2

Score

1.000

passed
design ran as designed
Cost
$0.243
Latency
54.6s
setup 0.0s, grading 0.0s
Turns
56
Tool calls
50
Subagents
3
Tokens in / out
145k / 11k
109,386 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 17.5s, finished, 11 tool calls, $0.043
Proposal 2 (state-first): 0.0s to 19.6s, finished, 11 tool calls, $0.045
Proposal 3 (outcome-first): 0.0s to 20.1s, finished, 14 tool calls, $0.051
Select: 20.1s to 34.2s, finished, 0 tool calls, $0.040
Execute: 34.2s to 54.6s, finished, 14 tool calls, $0.063

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

PhaseDurationCostTurnsToolsModel
Proposal 1 (policy-first)17.5s$0.0431211claude-sonnet-5-5
Proposal 2 (state-first)19.6s$0.0451211claude-sonnet-5-5
Proposal 3 (outcome-first)20.1s$0.0511514claude-sonnet-5-5
Select14.0s$0.04020claude-sonnet-5-5
Execute20.4s$0.0631514claude-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 propose-select --tasks support-medium --models claude-sonnet-5-5 --repeats 1

Score breakdown

  • passSurvivor carries the earliest SLA deadline of the merged set
    1.00weight 0.23

    sla_due_at=2026-03-17T12:40:00Z (expected 2026-03-17T12:40:00Z)

  • passBoth duplicate reports were merged into the earliest-created ticket
    1.00weight 0.23

    T-1053: status=merged, merged_into=T-1052; T-1054: status=merged, merged_into=T-1052

  • passThe engineering tracking ticket now points at the survivor
    1.00weight 0.15

    T-1061 links=[T-1052]

  • passSurvivor re-graded from the matrix for its widened impact scope
    1.00weight 0.15

    severity=sev2 (matrix says sev2)

  • passSurvivor assigned to the engineer on call for the owning rotation at its deadline
    1.00weight 0.15

    assignee=E-201 (expected E-201)

  • passEvery ticket folded in was read before the merge
    1.00weight 0.08

Penalties

No penalties were applied.

Scorer metadata

tool calls
50
failed tool calls
0
distinct tools
11

Final answer

I merged the three duplicate CRM search reports into T-1052, set it to sev2 and assigned it to E-201.

  • Merge: T-1053 and T-1054 are folded into T-1052, the earliest-created of the three. All three reported the same search 500 outage, and KB-23 says every failing-search report is one incident. The merge raised T-1052's impact to department.
  • Severity: I set T-1052 to sev2. The escalation matrix gives sev2 for crm at department scope.
  • Deadline: T-1052 now carries 2026-03-17T12:40Z, the earliest deadline committed in the set (T-1053's). This is earlier than T-1052's old 15:05Z. The duplicate rule requires it, and no requester's commitment moved earlier.
  • Assignee: T-1052 goes to E-201 Ines Kovac. She is the business-apps engineer on call at 12:40Z, when the deadline falls. This replaces E-202, who had T-1053 and is in the right rotation but isn't on call at that moment.
  • Links: T-1061 now links to T-1052 instead of T-1054. I checked this after the merge.
  • Comment: I left a comment on T-1052 recording the merge, the grading, the assignment and that it can't be closed while T-1061 is open.

Left alone

  • T-1055 (empty CSV export): KB-24 says it has a different root cause, and its requester says search works.
  • T-1061 (search index rebuild tracker): I did not merge it. It is an engineering tracking ticket rather than a customer report. It is the one open linked ticket, so T-1052 stays open until it is done.

I didn't escalate, because the task only asked for consolidation, grading and assignment. If you want it escalated, the matrix routes sev2 to platform-primary.

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.