[codex] ship ProofLoop live-prod reliability slate#182
Draft
HomenShum wants to merge 13 commits into
Draft
Conversation
|
The latest updates on your projects. Learn more about Vercel for GitHub.
|
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
Ships the ProofLoop live-prod reliability package: deterministic HMDA and Q3 benchmark lanes, Convex-env provider preflight receipts, non-retryable provider credit failures, cancel cleanup, browser receipt separation, SEC/XBRL benchmark claim tightening, and advanced finance / actuarial benchmark evidence.
Why
Live-prod ProofLoop was blocked by provider 402 retries, uncancelled long-running jobs, and a generic Q3 browser task that could stay
running 1/1000. The benchmark writeups also overclaimed SEC/XBRL scope and lacked committed reviewable evidence. A follow-up correction fixed the preflight source: production live provider readiness must check the production-serving Convex env, not only the local shell.What changed
proofloop:live:prod, while redacting raw key and balance values from committed receipts.proofloop:live:prodas the one-command production slate with automatic receipts.Deployment
zealous-goshawk-766synced withnpm run convex:deploy.dpl_2shHR2jQwCufMaghDG2Dt1VbFRhhis Ready and aliased tohttps://noderoom.liveandhttps://nodeagent.live.Validation
npm run buildnpm test -- --run tests/providerEgressPolicy.test.ts tests/agentJobsRuntime.test.ts tests/hmdaUnderwritingExecutor.test.ts tests/proofloopSupervisor.test.ts tests/secXbrlAudit.test.ts tests/q3VarianceExecutor.test.ts tests/agentJobsSource.test.tsnpm test -- --run tests/agentJobsSource.test.tsnpm run nodeagent:frame:smokenpm run omnigent:nodeagent:smokenpx tsc --noEmit --project convex/tsconfig.json --pretty falsenpm run benchmark:sec-xbrl:ingestnpm test -- --run tests/secXbrlAudit.test.tsnpm run benchmark:advanced-financenpm run proofloop:provider:preflight -- --key-source convex-env --convex-deployment zealous-goshawk-766 --min-balance-usd 1 --softnpm run proofloop:live:prod -- --continue-on-failureafter redeploy and again after Convex-env preflight correctionFinal live-prod receipt:
docs/eval/live-prod/live-prod-20260706T231618Z/suite-receipt.json.Known gated item
BTB live-prod is now correctly skipped when Convex-env provider preflight says
provider_insufficient_credits. The production-serving Convex env hasOPENROUTER_API_KEY, but the remaining credit bucket is below the configured$1threshold. With credits added, the same slate will run BTB instead of starting a doomed retry loop.