Repository navigation
test(live): answer the Current Debate prompt and pin Muse Spark 1.3 Contributor - #85
Conversation
…ontributor The live journeys were never updated for Current Debate (#72): Actions was out of minutes from 2026-09-27 to 2026-09-30, so the first real live-e2e run was on 2026-10-01. Five journeys stopped at "Which debate is this for?" or, in the Library, at a disabled Add Source button with no current debate. They now answer the prompt the way the full-stack specs do, and the Library journey starts a debate first. The live gate's default model is now meta/muse-spark-1.3-contributor ($0.10/$0.20 per million tokens, against $0.15/$0.50 for GLM 5.3 Flash). It supports every parameter the app sends. It is test-only and stays out of the app catalog. Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
|
The latest updates on your projects. Learn more about Vercel for GitHub.
|
|
Warning Review limit reachedEnable usage-based reviews in Billing to review now. Otherwise, wait until the next included review is available. Next included review available in 53 minutes. View limit detailsLimit details: You’ve used the included review currently available. Review configuration: ⚙️ Run configurationConfiguration used: Organization UI Review profile: CHILL Plan: Free Run ID: 📒 Files selected for processing (4)
Note 🎁 Summarized by CodeRabbit FreeYour organization is on the Free plan. CodeRabbit will generate a high-level summary and a walkthrough for each pull request. For a comprehensive line-by-line review, please upgrade your subscription to CodeRabbit Essentials by visiting https://app.coderabbit.ai/settings/billing. Comment |
Brings in #79 to #83 and #85. Conflict in .superset/db-gate.sh: both sides added a function to the readiness loop. The loop now waits for both `voice` and `account`. #82 replaced the "Which debate is this for?" dialog with the first-debate offer when no debate is current. e2e/voice-rounds.spec.ts still clicked the old button in two places, so it answers the new offer the way e2e/sparring.spec.ts does. Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Fixes the
live-e2efailure that every open PR andmainshow, and makes the live gate use the cheapest model.Why live-e2e fails
It isn't billing. From 2026-09-27 to 2026-09-30 every
live-e2erun died in 2 to 5 seconds without a runner, so the live suite never ran against Current Debate (#72). The runs on 2026-10-01 were the first real ones, and 5 of 8 journeys failed:The Edge "wall clock" terminations in those logs come from the browser closing when the tests timed out. The server wasn't hanging; the Attack call in the same run completed normally.
Changes
chooseForMotion(), the same way the full-stack specs do, and the Library journey callsstartDebate()first..superset/db-gate.shnow defaultsLIVE_MODELtometa/muse-spark-1.3-contributor. OpenRouter lists it at $0.10 / $0.20 per million input/output tokens, against $0.15 / $0.50 for the old default, GLM 5.3 Flash. Its endpoint supports every parameter the app sends:response_format, reasoning effort, temperature and max tokens. It's test-only and stays out of the app catalog, so users can't pick it. Meta is its only provider, so a Meta outage fails the gate instead of failing over.docs/testing.mdsays so.What still uses other models
typesafe/jev-1.13through Jev's own API. A chat model can't replace it.model-matrixdeliberately runs one case on every catalog model.Verification
docs:check, typecheck, ESLint one2e/liveandbash -non the gate script pass.live-e2ejob passed 8 of 8 journeys in 5.8 minutes withmeta/muse-spark-1.3-contributor.expectProviderCallsHealthychecks that every chat call used the pinned model and returned 200.browser-testsfailed once onnavigation-performance.spec.ts:118(route-skeleton timing). This PR doesn't touch that code, and the same test flaked and passed on rerun on feat(staging): admin staging switcher and staging sync procedure #81.gpt-6.1-sol, high): no actionable findings.🤖 Generated with Claude Code