From e0ad6d356f12cd9ccdefe786d151a403f65c4d72 Mon Sep 17 00:00:00 2001 From: Dotta Date: Thu, 3 Sep 2026 09:26:03 -0500 Subject: [PATCH] test(runner): retry proven breadth plan variance --- tests/runner-e2e/runner.spec.ts | 12 +++++++----- 1 file changed, 7 insertions(+), 5 deletions(-) diff --git a/tests/runner-e2e/runner.spec.ts b/tests/runner-e2e/runner.spec.ts index b3954504d6..a7c1d48a0b 100644 --- a/tests/runner-e2e/runner.spec.ts +++ b/tests/runner-e2e/runner.spec.ts @@ -1446,17 +1446,19 @@ for (const execution of executions) { } if ( execution.suite.id === "openrouter-model-breadth" && - execution.task.id === "question-resume-complete" && + (execution.task.id === "question-resume-complete" || + execution.task.id === "plan-approve-complete") && exactMessageMatcher?.kind === "message_exact" && selectedRuns.every((candidate) => candidate.status === "succeeded") && issue.status === "done" && finalRunMessage !== exactMessageMatcher.expected && record(finalRun.resultJson).summary === exactMessageMatcher.expected ) { - // A breadth model can occasionally satisfy the durable terminal - // contract while paraphrasing the visible final response. Preserve - // the exact persisted-message and DOM assertions, but classify this - // narrowly proven provider variance for one fresh-harness retry. + // An interactive breadth model can occasionally satisfy the durable + // terminal contract while paraphrasing the visible final response. + // Preserve the exact persisted-message and DOM assertions, but + // classify this narrowly proven provider variance for one + // fresh-harness retry. failureClassOverride = "provider_variance"; } const observedEnvironmentId =