Promote staging to main by stateofkate · Pull Request #1130 · abundant-ai/oddish

stateofkate · 2026-08-07T22:09:14Z

Caution

DO NOT USE THE MERGE BUTTON ON THIS PULL REQUEST.
THE BUTTON CREATES A NEW COMMIT AND BREAKS THE RELEASE MODEL.
COMMENT /promote TO COMPLETE THE PROMOTION.

Summary

Ships the quota lock removal, the QA accept/reject verdict rework, the network-shape e2e de-race, and /promote <sha> support in the promotion workflow.

Commits in this promotion

e43f64f8 ci(promote): support /promote <sha> and share the promotion checks (#1133)
8b7b2367 test(e2e): de-race the network-shape request counting (#1128)
fa608980 QA verdict: accept/reject labels, plain prompt, keep verdict until replaced, scoped rerun (#1120)
9f26989e fix(quotas): remove the org-wide advisory lock from trial submission (#1118)

Validation

Staging Deploy is green on the staging tip e43f64f8a1228767055eb76930fc9b0362cbe93a (run completed 2026-08-08T23:39:32Z). main is an ancestor of the tip, so the promotion fast-forwards cleanly.

Note

High Risk
Large cross-cutting release touching quota enforcement, task verdict persistence and constraints, promotion/git workflows, and core QA paths—incorrect verdict or quota behavior affects production task outcomes and spend controls.

Overview
This promotion advances main to the validated staging tip and ships four feature areas together.

Release & CI: Promotion Preflight and /promote now share verify_promotion_target.sh, resolving the target from /promote <sha>, the PR body's promotion-target pin, or the staging tip, with fast-forward and green Staging Deploy checks. PR preview's require-working-preview gate uses !cancelled() so superseded runs do not publish failing checks.

Quotas: Trial submission no longer takes org/user quota advisory locks; admission can briefly overshoot and the enforcement sweep cancels overage. Retries lock the task row before reserving {task_id}-{N}.

Task QA verdict: New oddish.core.verdict_state is the sole writer for verdict columns—published payloads stay visible while replacement QA runs; cancel restores, terminal failure clears. A DB check constraint enforces valid payload/status pairs. Verdicts use accept/reject (with is_good kept for compatibility); the judge prompt is rewritten for immediate-rejection rules. UI shows Accepted/Rejected, matrix cells use QA tone on pass/fail glyphs, task overview lists all version trials (with "elsewhere" chips), and Rerun verdict calls qa/backfill with force: false. GET /trials/.../trajectory/summary?refresh=true requires TASKS scope and bypasses schema-version cache.

E2E: Shared network-log.ts attributes requests by issue time and holds responses to beat StrictMode abort races in tasks/experiment network-shape specs.

^{Reviewed by Cursor Bugbot for commit 33fa252. Bugbot is set up for automated code reviews on this repo. Configure here.}

…1118) Co-authored-by: Claude <noreply@anthropic.com> Co-authored-by: kyle-compute <kylehvdbnl@gmail.com>

vercel · 2026-08-07T22:09:24Z

The latest updates on your projects. Learn more about Vercel for GitHub.

Project	Deployment	Actions	Updated (UTC)
oddish-app	Ready	Preview	Aug 10, 2026 10:29pm
oddish-app (staging)	Ready	Preview	Aug 10, 2026 10:29pm

…placed, scoped rerun (#1120) Co-authored-by: Claude <noreply@anthropic.com> Co-authored-by: kyle-compute <kylehvdbnl@gmail.com>

cursor

Risk: high. Left a non-blocking comment: Cursor Bugbot finished as skipped and left an unresolved high-severity finding on quota-cancel verdict handling, so this is not approved. Assigned reviewers for human review of this billing/quota promotion.

_{Sent by Cursor Approval Agent: Pull Request Router and Approver}

…1133)

…R body (#1134)

) Co-authored-by: Claude Opus 5 <noreply@anthropic.com> Co-authored-by: kyle-compute <kylehvdbnl@gmail.com>

cursor

Cursor Bugbot has reviewed your changes using high effort and found 1 potential issue.

There are 2 total unresolved issues (including 1 from previous review).

^{❌ Bugbot Autofix is OFF. To automatically fix reported issues with cloud agents, enable autofix in the Cursor dashboard.}

^{Reviewed by Cursor Bugbot for commit 319fe01. Configure here.}

Co-authored-by: dot-agi <ps4534@nyu.edu>

kyle-compute · 2026-08-10T23:04:21Z

/promote 33fa252

github-actions · 2026-08-10T23:04:36Z

Promoted. main now points at 33fa252eeebbe52003a14e48b46b73a0ed0d45c3, the same commit as staging.

fix(quotas): remove the org-wide advisory lock from trial submission (#…

9f26989

…1118) Co-authored-by: Claude <noreply@anthropic.com> Co-authored-by: kyle-compute <kylehvdbnl@gmail.com>

github-actions Bot deployed to Preview August 7, 2026 22:09 View deployment

QA verdict: accept/reject labels, plain prompt, keep verdict until re…

fa60898

…placed, scoped rerun (#1120) Co-authored-by: Claude <noreply@anthropic.com> Co-authored-by: kyle-compute <kylehvdbnl@gmail.com>

kyle-compute temporarily deployed to staging August 7, 2026 22:13 — with GitHub Actions Inactive

github-actions Bot deployed to Preview August 7, 2026 22:13 View deployment

cursor Bot reviewed Aug 7, 2026

View reviewed changes

Comment thread oddish/src/oddish/core/quota_enforcement.py

cursor Bot requested review from RishiDesai and charlesyhuang August 7, 2026 22:17

cursor Bot reviewed Aug 7, 2026

View reviewed changes

test(e2e): de-race the network-shape request counting (#1128)

8b7b236

kyle-compute temporarily deployed to staging August 8, 2026 00:18 — with GitHub Actions Inactive

github-actions Bot deployed to Preview August 8, 2026 00:18 View deployment

kyle-compute mentioned this pull request Aug 8, 2026

ci(promote): support /promote <sha> and share the promotion checks #1133

Merged

ci(promote): support /promote <sha> and share the promotion checks (#…

e43f64f

…1133)

kyle-compute temporarily deployed to staging August 8, 2026 23:37 — with GitHub Actions Inactive

github-actions Bot deployed to Preview August 8, 2026 23:37 View deployment

ci(promote): pin bare /promote to the sha recorded in the promotion P…

33a8e4e

…R body (#1134)

kyle-compute temporarily deployed to staging August 9, 2026 00:37 — with GitHub Actions Inactive

github-actions Bot deployed to Preview August 9, 2026 00:37 View deployment

cursor Bot reviewed Aug 9, 2026

View reviewed changes

Comment thread oddish/src/oddish/workers/queue/qa_handler.py

fix(summary): allow a stored trajectory summary to be regenerated (#1112

319fe01

) Co-authored-by: Claude Opus 5 <noreply@anthropic.com> Co-authored-by: kyle-compute <kylehvdbnl@gmail.com>

stateofkate temporarily deployed to staging August 10, 2026 21:00 — with GitHub Actions Inactive

github-actions Bot deployed to Preview August 10, 2026 21:00 View deployment

cursor Bot reviewed Aug 10, 2026

View reviewed changes

Comment thread oddish/src/oddish/core/verdict_sync.py Outdated

fix(qa): preserve verdict on quota cancellation (#1138)

353be62

kyle-compute temporarily deployed to staging August 10, 2026 21:29 — with GitHub Actions Inactive

github-actions Bot deployed to Preview August 10, 2026 21:29 View deployment

fix: preserve verdict state and skip cancelled preview gates (#1139)

7f06406

Co-authored-by: dot-agi <ps4534@nyu.edu>

kyle-compute temporarily deployed to staging August 10, 2026 22:11 — with GitHub Actions Inactive

github-actions Bot deployed to Preview August 10, 2026 22:11 View deployment

feat(ui): show expected good failures as a green ✗ (#1140)

33fa252

kyle-compute temporarily deployed to staging August 10, 2026 22:28 — with GitHub Actions Inactive

github-actions Bot deployed to Preview August 10, 2026 22:28 View deployment

dot-agi merged commit 33fa252 into main Aug 10, 2026
26 checks passed

railway-app Bot temporarily deployed to pleasing-unity / production August 10, 2026 23:04 Inactive

vercel Bot deployed to Production August 10, 2026 23:05 View deployment

Navigation Menu

Uh oh!

Promote staging to main - #1130

Promote staging to main#1130
dot-agi merged 9 commits into
mainfrom
staging

stateofkate commented Aug 7, 2026 •

edited by cursor Bot

Loading

Uh oh!

vercel Bot commented Aug 7, 2026 •

edited

Loading

Uh oh!

Uh oh!

cursor Bot left a comment

Uh oh!

Uh oh!

cursor Bot left a comment

Uh oh!

Uh oh!

kyle-compute commented Aug 10, 2026

Uh oh!

github-actions Bot commented Aug 10, 2026

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

4 participants

Uh oh!

Conversation

stateofkate commented Aug 7, 2026 • edited by cursor Bot Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Summary

Commits in this promotion

Validation

Uh oh!

vercel Bot commented Aug 7, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

Uh oh!

cursor Bot left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

cursor Bot left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

kyle-compute commented Aug 10, 2026

Uh oh!

github-actions Bot commented Aug 10, 2026

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

4 participants

stateofkate commented Aug 7, 2026 •

edited by cursor Bot

Loading

vercel Bot commented Aug 7, 2026 •

edited

Loading