Sutura Case Lab

Deterministic flaky failure

Recorded live result Flaky, no patch

No patch: the failure was not consistent

The failure did not reproduce consistently, so any patch would be guesswork.

Historical recorded result · A timing race fails two runs in five. Sutura must classify the flake and refuse to invent a patch. · subject 3b6152d9d72beebd7dcbaaa4045c94d640456adb · 2026-10-27T07:22:31.298Z

Stabilize the test before asking for a repair.

This result is a recorded live benchmark evaluation of the released Sutura version, not a run started from this page.

Outcome matches the expected outcome.

Failed commit and CI evidence

Nano diagnosis and confidence

Failure class
test-assertion
Confidence
49%
Signals
  • AssertionError
  • expected false to be true
  • mechanical:test-assertion
  • signature:\bexpected\b[^\n]*\b(?:to|but)\b[^\n]*\b(?:received|be|equal|match)\b
  • llm-command-mismatch
  • llm-excerpt-ungrounded

Clean audit branch and Ultra verdict

Not run. No candidate survived for adversarial audit.

Final outcome

Flaky, no patch Recorded live result

Sutura never merges a generated repair. A flaky, no patch result still needs human review.

Check a patch this page did not produce

This run did not record an exact failing commit and policy commit, so both are shown as placeholders. Verification is always tied to exact commits. A patch cannot ask for a more permissive policy by carrying one, and no green log is accepted in place of execution.

sutura verify \
  --case-dir <checkout> \
  --source-sha <failing commit> \
  --policy-base-sha <trusted policy commit> \
  --candidate-diff <patch file> \
  --failing-command diagnosed \
  --format json

The same route runs as a GitHub Action and needs no repository write access. Uploading a patch stays a command-line and Action route; this page never runs one.

Execution and cost detail
Release
v0.4.1 · Action 3b6152d9d72beebd7dcbaaa4045c94d640456adb
Controller
3b6152d9d72beebd7dcbaaa4045c94d640456adb
Request
recorded-flaky-failure · 2026-10-30T06:58:46.084Z

ConTree search tree and branch status

Triage: intermittent · reproduced 2/5 · stop reason maximum-attempts · method sprt-p20-p80-a05-b05-v1

No repair search branches were opened.

Super proposals and candidate patches

No candidate patch was proposed.

Rejected patches and rejection reasons

No candidate was rejected.

Token cost, provider cost, latency, and sandbox operations

Inference cost
USD 0.000264
Sandbox cost
USD 0.074841
Elapsed
100.0 s
CPU time
6.92 s
Peak memory
151 MiB
Sandbox stages
16 (7 with an operation id)
RoleCallsInput tokensOutput tokensCost
nano12076580USD 0.000264

Recorded from docs/demo/placebo-v0.4.1-live-2026-10-27.json (result hash d8eb9f18c578…) at 2026-10-27T07:22:31.298Z, subject 3b6152d9d72beebd7dcbaaa4045c94d640456adb.

Result hash eeac1f3c317bb1e2fc940db53fc52cc14d135ae96b1ab62998d8466096e1a7a2