Fix Codex synthetic usage in status without local OpenAI profiles by brokemac79 · Pull Request #92520 · openclaw/openclaw

brokemac79 · 2026-06-12T19:27:05Z

Summary

route /status Codex synthetic usage through the Codex app-server marker even when no local OpenAI OAuth/token profile label exists
teach shared status --usage to add the Codex synthetic OpenAI usage snapshot only for configured OpenAI Codex runtime routes whose effective auth label is not API-key-backed
keep API-key/OpenClaw-pinned status contexts from probing Codex synthetic usage, with regression coverage for no-profile, API-key-backed, and OpenClaw-pinned cases

Linked context

Issue: [Bug]: /status usage missing for OpenAI Codex synthetic auth after 2026.6.6 #92506
ClawSweeper intake: source-repro, queueable-fix, fix-shape-clear
Current PR head: 08e19d72ac
Upstream contract checked locally in sibling ../codex: account/rateLimits/read maps to GetAccountRateLimits and returns Codex rate-limit snapshots used by the existing Codex provider usage hook

Real behavior proof

Behavior or issue addressed: status --usage should show real Codex/OpenAI usage windows when the default OpenAI model is effectively running on the Codex runtime, even if the status process does not have a local OpenAI OAuth/token profile label. Current 2026.6.6 omitted the Codex/OpenAI usage line.
Real environment tested: production OpenClaw VPS polymarket-mc, Ubuntu arm64, prod OpenClaw home/config/gateway, gateway still running from installed OpenClaw 2026.6.6. After-fix command was run from an isolated PR checkout at ~/oc-proof-92520, commit 220380b3, without replacing or restarting the prod gateway service.
Current-head delta after proof: 08e19d72ac only adds the shared CLI effective-auth guard so API-key-backed OpenAI Codex runtime routes do not receive the synthetic app-server quota query; it does not change the proven no-profile app-server path.
Exact steps or command run after this patch:

# before, installed production CLI
openclaw --version
openclaw status --usage

# after, isolated PR checkout against the same real ~/.openclaw state and gateway
cd ~/oc-proof-92520
node openclaw.mjs --version
node openclaw.mjs status --usage

Evidence after fix:

=== BEFORE installed openclaw version ===
OpenClaw 2026.6.6 (8c802aa)
=== BEFORE installed openclaw status --usage ===
Sessions
... default gpt-5.5 ... Runtime OpenAI Codex ...

Usage
Usage:
  Claude: HTTP 403: OAuth token <redacted> not meet scope requirement user:profile

=== AFTER PR checkout openclaw version ===
OpenClaw 2026.6.2 (220380b)
=== AFTER PR checkout openclaw status --usage ===
Overview
  Git                  fix/92506-codex-synthetic-usage-status · @ 220380b3
  Gateway              local · ws://127.0.0.1:18789 ... app 2026.6.6 linux 6.17.0-1011-oracle
  Sessions             255 active · default gpt-5.5 (200k ctx) · 2 stores

Sessions
... default gpt-5.5 ... Runtime OpenAI Codex ...

Usage
Usage:
  Claude: HTTP 403: OAuth token <redacted> not meet scope requirement user:profile
  OpenAI (pro (<redacted> credits))
    5h: 86% left · resets 58m
    Week: 86% left · resets 5d 4h

Observed result after fix: the same VPS/prod config that previously rendered only the Claude usage error now renders real Codex/OpenAI usage from the app-server route, including both 5h and Week windows. The gateway stayed on the installed prod 2026.6.6 service; only the CLI process came from the isolated PR checkout.
What was not tested: I did not post Telegram /status screenshots because the real CLI status --usage output exercises the requested status usage path and ClawSweeper accepted copied live output as sufficient. The after command did not replace or restart the production gateway.

Supplemental tests and validation

node scripts/run-vitest.mjs src/commands/status-runtime-shared.test.ts -- --reporter=verbose --pool=forks --maxWorkers=1 --fileParallelism=false
node scripts/run-vitest.mjs src/auto-reply/reply/commands-status.test.ts src/commands/status-runtime-shared.test.ts src/commands/status-json.test.ts src/infra/provider-usage.auth.plugin.test.ts src/infra/provider-usage.load.plugin.test.ts -- --reporter=verbose --pool=forks --maxWorkers=1 --fileParallelism=false
node scripts/run-vitest.mjs extensions/codex/provider.test.ts extensions/codex/src/app-server/rate-limits.test.ts -- --reporter=verbose --pool=forks --maxWorkers=1 --fileParallelism=false
node scripts/run-tsgo.mjs -p tsconfig.core.json --noEmit --incremental false
node scripts/run-tsgo.mjs -p tsconfig.extensions.json --noEmit --incremental false
git diff --check
codex -c service_tier='''fast''' review --base origin/main -> no actionable regressions found before the current-head API-key guard follow-up

Risk checklist

No merge, land, or automerge performed
No changelog added; scoped status/runtime bugfix with regression tests
Current-head CI has one unrelated cleanup-flake failure in checks-node-agentic-control-plane-agent-chat: ENOTEMPTY while removing /tmp/openclaw-gw-*; GitHub permissions did not allow me to rerun it
ClawSweeper re-review requested after adding real VPS proof and the API-key guard follow-up

clawsweeper · 2026-06-12T19:28:59Z

Codex review: needs changes before merge. Reviewed June 25, 2026, 8:27 PM ET / 00:27 UTC.

Summary
The PR adds shared Codex synthetic OpenAI usage routing for status output and regression coverage for no-profile, API-key-backed, and OpenClaw-pinned status contexts.

PR surface: Source +111, Tests +236. Total +347 across 5 files.

Reproducibility: yes. The linked issue and PR body give a concrete live 2026.6.6 openclaw status --usage before/after path; I did not rerun the VPS flow, but the current-main source path still matches the missing no-profile case.

Review metrics: 1 noteworthy metric.

Synthetic usage probes: 1 conditional OpenAI query added. The PR adds a second usage collection path for OpenAI/Codex routes, so merge precedence directly affects what users see in status output.

Root-cause cluster
Relationship: fixed_by_candidate
Canonical: #92506
Summary: This PR is the candidate fix for the open no-local-profile Codex synthetic usage issue; prior related work restored part of the status usage path but did not cover this remaining case.

Members:

canonical: [Bug]: /status usage missing for OpenAI Codex synthetic auth after 2026.6.6 #92506 - The PR body says it closes this open issue, and the issue describes the exact no-local-profile OpenAI/Codex synthetic status usage regression.
partial_overlap: fix(status): restore Codex synthetic usage line #91709 - That merged PR restored the earlier Codex synthetic usage line, but the follow-up issue states 2026.6.6 still missed the no-local-profile case.
superseded: [Bug]: Codex/OpenAI usage line disappears from status after 2026.6.5-beta.6 #91694 - That earlier issue was closed by the merged synthetic usage restoration and is now historical context for the remaining no-profile regression.
adjacent_distinct: fix(status): use legacy Codex OAuth profiles for OpenAI usage #90356 - That open PR concerns legacy openai-codex OAuth profile compatibility, while this PR addresses synthetic Codex app-server usage without local OpenAI profiles.
partial_overlap: Status line reports api-key (env: OPENAI_API_KEY) even when codex runtime is authenticated via ChatGPT #91099 - That issue is another status/auth-provider reporting problem around Codex ChatGPT auth versus API-key labels, but it is not the same missing usage snapshot root cause.

Proposal only: this assessment does not dispatch repair, suppress jobs, mutate sibling items, close, or merge anything.

Merge readiness
Overall: 🦐 gold shrimp
Proof: 🦞 diamond lobster
Patch quality: 🦐 gold shrimp
Result: needs maintainer review before merge.

Overall follows the weaker of proof and patch quality, so missing proof can cap an otherwise strong patch.

Rank-up moves:

[P2] Fix the empty synthetic snapshot merge preference and add a focused regression test.

Mantis proof suggestion
A native Telegram /status recording would directly show the visible chat output this shared status path affects. A maintainer can ask Mantis to capture proof by posting this exact PR comment:

@openclaw-mantis telegram desktop proof: verify /status shows OpenAI/Codex usage for a Codex runtime with no local OpenAI profile.

Risk before merge

[P1] Shared CLI/JSON status can replace a useful existing OpenAI usage snapshot with an empty successful synthetic Codex snapshot, making status show OpenAI: no data or omit meaningful windows for existing users.
[P1] The change is still auth-provider-sensitive because it adds an extra Codex app-server rate-limit query based on effective credential type and runtime routing.

Maintainer options:

Preserve meaningful base usage (recommended)
Change the merge helper so an empty successful synthetic OpenAI snapshot cannot overwrite existing OpenAI windows or summary, then add a regression in status-runtime-shared.test.ts.
Accept status no-data risk
Maintainers could accept that synthetic Codex usage always wins, but that would knowingly risk hiding existing useful status usage output.

Copy recommended automerge instruction

@clawsweeper automerge

Special instructions:
Preserve useful existing OpenAI usage snapshots when the Codex synthetic snapshot succeeds with no windows or summary; add a status-runtime-shared regression where the base OpenAI snapshot has windows and the synthetic OpenAI result is empty.

Next step before merge

[P2] A narrow, file-local merge-preference repair remains and can be attempted without a product decision.

Security
Cleared: No dependency, workflow, lockfile, or new secret-storage surface was added; the security-sensitive change reuses the existing Codex app-server auth marker for a status-only rate-limit query.

Review findings

[P2] Preserve useful usage snapshots when synthetic Codex is empty — src/status/codex-synthetic-usage.ts:68-69

Review details

Best possible solution:

Land after mergeUsageSummaries preserves meaningful existing OpenAI snapshots unless the synthetic Codex result has windows or a summary, with a focused regression test for that case.

Do we have a high-confidence way to reproduce the issue?

Yes. The linked issue and PR body give a concrete live 2026.6.6 openclaw status --usage before/after path; I did not rerun the VPS flow, but the current-main source path still matches the missing no-profile case.

Is this the best way to solve the issue?

No. The synthetic Codex route is the right layer, but the current merge preference is not the best fix until empty synthetic snapshots stop replacing useful existing OpenAI usage data.

Full review comments:

[P2] Preserve useful usage snapshots when synthetic Codex is empty — src/status/codex-synthetic-usage.ts:68-69
mergeUsageSummaries currently lets any no-error synthetic OpenAI snapshot replace the base OpenAI entry. Because loadProviderUsageSummary keeps successful empty snapshots and the Codex parser can return { windows: [] } without an error when no recognizable rate-limit payload is present, a user who already has a useful OpenAI usage snapshot can regress to OpenAI: no data. Prefer the extra snapshot only when it has windows/summary, or when the existing entry is missing, errored, or empty.
Confidence: 0.91

Overall correctness: patch is incorrect
Overall confidence: 0.9

AGENTS.md: found and applied where relevant.

Codex review notes: model internal, reasoning high; reviewed against dd0e4f6e61b8.

Label changes

Label changes:

add rating: 🦐 gold shrimp: Overall readiness is 🦐 gold shrimp; proof is 🦞 diamond lobster and patch quality is 🦐 gold shrimp.
remove rating: 🧂 unranked krab: Current PR rating is rating: 🦐 gold shrimp, so this older rating label is no longer current.

Label justifications:

P2: This is a normal-priority status/auth-provider visibility bug with limited blast radius: model execution still works, but usage visibility can be wrong.
merge-risk: 🚨 auth-provider: Merging changes provider-usage routing for OpenAI/Codex auth paths and can hide or alter usage snapshots based on credential detection.
rating: 🦐 gold shrimp: Overall readiness is 🦐 gold shrimp; proof is 🦞 diamond lobster and patch quality is 🦐 gold shrimp.
status: ⏳ waiting on author: ClawSweeper has contributor-facing work open and is waiting for author action. Sufficient (live_output): The PR body includes redacted copied live VPS before/after CLI output showing OpenAI/Codex usage windows after the patch; the later guard changes do not remove the proven no-profile path.
proof: sufficient: Contributor real behavior proof is sufficient. The PR body includes redacted copied live VPS before/after CLI output showing OpenAI/Codex usage windows after the patch; the later guard changes do not remove the proven no-profile path.
mantis: telegram-visible-proof: Mantis should capture Telegram visible proof. The changed status usage line is visible in Telegram /status, so a short Telegram Desktop proof would be useful even though the CLI live proof is sufficient for the core path.

Evidence reviewed

PR surface:

Source +111, Tests +236. Total +347 across 5 files.

View PR surface stats

Area	Files	Added	Removed	Net
Source	3	152	41	+111
Tests	2	349	113	+236
Docs	0	0	0	0
Config	0	0	0	0
Generated	0	0	0	0
Other	0	0	0	0
Total	5	501	154	+347

Acceptance criteria:

[P1] node scripts/run-vitest.mjs src/commands/status-runtime-shared.test.ts -- --reporter=verbose --pool=forks --maxWorkers=1 --fileParallelism=false.
[P1] node scripts/run-vitest.mjs src/auto-reply/reply/commands-status.test.ts src/commands/status-runtime-shared.test.ts src/commands/status-json.test.ts -- --reporter=verbose --pool=forks --maxWorkers=1 --fileParallelism=false.
[P1] node scripts/run-tsgo.mjs -p tsconfig.core.json --noEmit --incremental false.
[P1] git diff --check.

What I checked:

Current main still has the reported gap: Current main only enables chat status Codex synthetic usage when the resolved usage credential type is oauth or token, so a no-local-profile synthetic Codex route can still skip the OpenAI usage line. (src/status/status-text.ts:455, dd0e4f6e61b8)
PR adds a second synthetic usage read: PR head loads the normal provider usage summary first, then conditionally loads a second OpenAI summary through Codex synthetic auth and merges the two summaries. (src/commands/status-runtime-shared.ts:121, afa4474595dd)
Merge preference can overwrite useful base data: The PR helper treats any successful no-error extra snapshot as preferred, so an empty synthetic OpenAI snapshot can replace a base OpenAI snapshot that had windows or a summary. (src/status/codex-synthetic-usage.ts:68, afa4474595dd)
Provider usage keeps successful empty snapshots: The existing provider usage loader keeps snapshots with no error even when they have no windows or summary, which makes the PR's empty-success overwrite path reachable. (src/infra/provider-usage.load.ts:151, dd0e4f6e61b8)
Codex usage parser can return empty success: The Codex app-server usage parser returns an OpenAI snapshot with whatever windows it can extract and no error field; if no recognizable rate-limit snapshot is present, that is a successful empty provider snapshot. (extensions/codex/src/app-server/rate-limits.ts:196, dd0e4f6e61b8)
Upstream Codex contract checked: Sibling Codex source defines account/rateLimits/read as a client request returning GetAccountRateLimitsResponse, and its app-server tests show the endpoint requires ChatGPT auth and returns primary/secondary rate-limit windows. (../codex/codex-rs/app-server-protocol/schema/typescript/ClientRequest.ts:91, 5044062704af)

Likely related people:

brokemac79: Authored merged PR fix(status): restore Codex synthetic usage line #91709, which restored the Codex synthetic usage status line that this PR follows up. (role: previous fix author; confidence: high; commits: 1893a0727a37; files: src/status/status-text.ts, extensions/codex/provider.ts, extensions/codex/src/app-server/rate-limits.ts)
obviyus: Assigned on this PR and authored the recent runtime-authority status refactor that current status usage selection builds on. (role: recent status area contributor; confidence: medium; commits: 29670c13f671; files: src/status/status-text.ts)
zhangguiping-xydt: Authored merged PR fix(status): usage follows session model after switch #93384 for status usage following the session-selected model, directly adjacent to this PR's status usage selection. (role: recent status usage contributor; confidence: high; commits: bead84f0ee6e; files: src/status/status-text.ts)
Peter Steinberger: Introduced and tightened the Codex app-server extension/provider lifecycle that the synthetic usage hook relies on. (role: Codex app-server feature history; confidence: medium; commits: dd26e8c44d4e, 31a0b7bd42a5, 659bcc5e5b59; files: extensions/codex/provider.ts)

What the crustacean ranks mean

🦀 challenger crab: rare, exceptional readiness with strong proof, clean implementation, and convincing validation.
🦞 diamond lobster: very strong readiness with only minor maintainer review expected.
🐚 platinum hermit: good normal PR, likely mergeable with ordinary maintainer review.
🦐 gold shrimp: useful signal, but proof or patch confidence is still limited.
🦪 silver shellfish: thin signal; proof, validation, or implementation needs work.
🧂 unranked krab: not merge-ready because proof is missing/unusable or there are serious correctness or safety concerns.
🌊 off-meta tidepool: rating does not apply to this item.

Shiny media proof means a screenshot, video, or linked artifact directly shows the changed behavior. Runtime, network, CSP, and security claims still need visible diagnostics.

How this review workflow works

ClawSweeper keeps one durable marker-backed review comment per issue or PR.
Re-runs edit this comment so the latest verdict, findings, and automation markers stay together instead of adding duplicate bot comments.
A fresh review can be triggered by eligible @clawsweeper re-review comments, exact-item GitHub events, scheduled/background review runs, or manual workflow dispatch.
PR/issue authors and users with repository write access can comment @clawsweeper re-review or @clawsweeper re-run on an open PR or issue to request a fresh review only.
Maintainers can also comment @clawsweeper review to request a fresh review only.
Fresh-review commands do not start repair, autofix, rebase, CI repair, or automerge.
Maintainer-only repair and merge flows require explicit commands such as @clawsweeper autofix, @clawsweeper automerge, @clawsweeper fix ci, or @clawsweeper address review.
Maintainers can comment @clawsweeper explain to ask for more context, or @clawsweeper stop to stop active automation.

brokemac79 · 2026-06-12T20:00:29Z

@clawsweeper re-review

Added real production VPS before/after proof to the PR body. The before command is installed OpenClaw 2026.6.6 on polymarket-mc showing OpenAI Codex sessions but no OpenAI/Codex usage line; the after command is the isolated PR checkout at 220380b3 against the same real ~/.openclaw state and gateway, showing OpenAI usage with both 5h and Week windows.

clawsweeper · 2026-06-12T20:00:32Z

🦞🧹
ClawSweeper re-review requested.

I asked ClawSweeper to review this item again.
Action: item re-review queued (workflow sweep.yml, event repository_dispatch).
Result: the existing ClawSweeper review comment will be edited in place when the review finishes.

Re-review progress:

State: Complete
Detail: The targeted re-review finished, the durable review comment was updated, and the synced verdict was routed.
Run: https://github.com/openclaw/clawsweeper/actions/runs/27439808305
Updated: 2026-06-12T20:07:41.117Z

brokemac79 · 2026-06-12T20:30:00Z

@clawsweeper re-review

Addressed the effective-auth routing finding in current head 08e19d72ac: shared CLI status --usage now checks the resolved auth label and skips the synthetic Codex app-server usage query for API-key-backed OpenAI Codex runtime routes. Added a focused status-runtime-shared regression for that API-key case while preserving the no-profile Codex synthetic usage path and OpenClaw-pinned exclusion.

Validation run after the fix:

node scripts/run-vitest.mjs src/commands/status-runtime-shared.test.ts -- --reporter=verbose --pool=forks --maxWorkers=1 --fileParallelism=false
node scripts/run-vitest.mjs src/auto-reply/reply/commands-status.test.ts src/commands/status-runtime-shared.test.ts src/commands/status-json.test.ts src/infra/provider-usage.auth.plugin.test.ts src/infra/provider-usage.load.plugin.test.ts -- --reporter=verbose --pool=forks --maxWorkers=1 --fileParallelism=false
node scripts/run-vitest.mjs extensions/codex/provider.test.ts extensions/codex/src/app-server/rate-limits.test.ts -- --reporter=verbose --pool=forks --maxWorkers=1 --fileParallelism=false
node scripts/run-tsgo.mjs -p tsconfig.core.json --noEmit --incremental false
node scripts/run-tsgo.mjs -p tsconfig.extensions.json --noEmit --incremental false
git diff --check

Real VPS before/after proof remains in the PR body and was already marked sufficient; current-head delta only narrows eligibility for API-key-backed routes. One current CI shard is failing with an unrelated temp cleanup ENOTEMPTY in checks-node-agentic-control-plane-agent-chat; neither the GitHub app nor gh has permission to rerun it from my account.

clawsweeper · 2026-06-12T20:30:03Z

🦞🧹
ClawSweeper re-review requested.

I asked ClawSweeper to review this item again.
Action: item re-review queued (workflow sweep.yml, event repository_dispatch).
Result: the existing ClawSweeper review comment will be edited in place when the review finishes.

Re-review progress:

State: Superseded
Detail: A newer re-review for this item started before this run finished, so GitHub cancelled this older run. Check the latest ClawSweeper run for the current result.
Run: https://github.com/openclaw/clawsweeper/actions/runs/27441303852
Updated: 2026-06-12T20:32:40.584Z

obviyus · 2026-06-26T00:37:50Z

Landed via rebase onto main.

Scoped tests: node scripts/run-vitest.mjs src/commands/status-runtime-shared.test.ts src/auto-reply/reply/commands-status.test.ts src/commands/status-json.test.ts -- --reporter=verbose --pool=forks --maxWorkers=1 --fileParallelism=false; node scripts/run-tsgo.mjs -p tsconfig.core.json --noEmit --incremental false; git diff --check origin/main...HEAD; node scripts/check-no-conflict-markers.mjs; .agents/skills/autoreview/scripts/autoreview --mode branch --base origin/main
Changelog: not edited; normal bugfix PR, release generation owns changelog.
Land commit: 2b2e93a
Merge commit: a0b3977

Thanks @brokemac79!

openclaw-barnacle Bot added commands Command implementations size: L triage: needs-real-behavior-proof Candidate: external PR needs after-fix proof from a real setup. labels Jun 12, 2026

openclaw-barnacle Bot added proof: supplied External PR includes structured after-fix real behavior proof. and removed triage: needs-real-behavior-proof Candidate: external PR needs after-fix proof from a real setup. labels Jun 12, 2026

clawsweeper Bot had a problem deploying to qa-live-shared June 12, 2026 19:37 Failure

clawsweeper Bot added the P2 Normal backlog priority with limited blast radius. label Jun 12, 2026

brokemac79 force-pushed the fix/92506-codex-synthetic-usage-status branch from 220380b to 08e19d7 Compare June 12, 2026 20:15

openclaw-barnacle Bot removed the proof: sufficient ClawSweeper judged the real behavior proof convincing. label Jun 12, 2026

clawsweeper Bot added the proof: sufficient ClawSweeper judged the real behavior proof convincing. label Jun 12, 2026

clawsweeper Bot added rating: 🦐 gold shrimp Decent PR readiness signal, but merge confidence is limited. status: ⏳ waiting on author ClawSweeper has contributor-facing work open and is waiting for author action. labels Jun 19, 2026

clawsweeper Bot mentioned this pull request Jun 19, 2026

[Bug]: /status usage missing for OpenAI Codex synthetic auth after 2026.6.6 #92506

Closed

clawsweeper Bot added the mantis: telegram-visible-proof Mantis should capture Telegram visible proof. label Jun 20, 2026

clawsweeper Bot had a problem deploying to qa-live-shared June 20, 2026 18:46 Failure

brokemac79 force-pushed the fix/92506-codex-synthetic-usage-status branch from 830dda2 to da9f57f Compare June 21, 2026 12:47

openclaw-barnacle Bot removed the scripts Repository scripts label Jun 21, 2026

clawsweeper Bot mentioned this pull request Jun 21, 2026

Status line reports api-key (env: OPENAI_API_KEY) even when codex runtime is authenticated via ChatGPT #91099

Closed

obviyus self-assigned this Jun 25, 2026

obviyus force-pushed the fix/92506-codex-synthetic-usage-status branch from afa4474 to 89e98a2 Compare June 26, 2026 00:26

clawsweeper Bot added rating: 🦐 gold shrimp Decent PR readiness signal, but merge confidence is limited. and removed rating: 🧂 unranked krab Not merge-ready due to missing proof or serious correctness/safety concerns. labels Jun 26, 2026

fix(status): restore Codex synthetic usage in status

2b2e93a

obviyus force-pushed the fix/92506-codex-synthetic-usage-status branch from 89e98a2 to 2b2e93a Compare June 26, 2026 00:31

obviyus merged commit a0b3977 into openclaw:main Jun 26, 2026
94 checks passed

Haderach-Ram mentioned this pull request Jun 26, 2026

Ecosystem Digest — 2026-06-26 Haderach-Ram/openclaw-radar#62

Open

clawsweeper Bot mentioned this pull request Jun 28, 2026

fix(infra): wrap provider auth resolution in timeout for status --usage --json #74185

Closed

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Uh oh!

Uh oh!

Fix Codex synthetic usage in status without local OpenAI profiles#92520

Fix Codex synthetic usage in status without local OpenAI profiles#92520
obviyus merged 1 commit into
openclaw:mainfrom
brokemac79:fix/92506-codex-synthetic-usage-status

brokemac79 commented Jun 12, 2026 •

edited

Loading

Uh oh!

clawsweeper Bot commented Jun 12, 2026 •

edited

Loading

Uh oh!

brokemac79 commented Jun 12, 2026

Uh oh!

clawsweeper Bot commented Jun 12, 2026 •

edited

Loading

Uh oh!

brokemac79 commented Jun 12, 2026

Uh oh!

clawsweeper Bot commented Jun 12, 2026 •

edited

Loading

Uh oh!

Uh oh!

obviyus commented Jun 26, 2026

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

Uh oh!

Uh oh!

Conversation

brokemac79 commented Jun 12, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Summary

Linked context

Real behavior proof

Supplemental tests and validation

Risk checklist

Uh oh!

clawsweeper Bot commented Jun 12, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

brokemac79 commented Jun 12, 2026

Uh oh!

clawsweeper Bot commented Jun 12, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

brokemac79 commented Jun 12, 2026

Uh oh!

clawsweeper Bot commented Jun 12, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

Uh oh!

obviyus commented Jun 26, 2026

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

brokemac79 commented Jun 12, 2026 •

edited

Loading

clawsweeper Bot commented Jun 12, 2026 •

edited

Loading

clawsweeper Bot commented Jun 12, 2026 •

edited

Loading

clawsweeper Bot commented Jun 12, 2026 •

edited

Loading