feat(agents): allow authorized private credential handoffs by obviyus · Pull Request #139196 · openclaw/openclaw

obviyus · 2026-09-05T15:52:04Z

What Problem This Solves

Credential guidance prohibited every authentication code in transcripts, conflicting with authorized private sign-in handoffs. The live baseline already completed the tested sign-in; this is a requested policy change rather than a reproduced live refusal.

Why This Change Was Made

Shared guidance permits matching short-lived codes only in the requester's private conversation. Group acknowledgement follows successful private delivery. Matching pending flows validate callbacks and confirm completion before success is reported. Passwords, reusable tokens, keys, cookies, and recovery codes still require protected entry.

User Impact

A sign-in request authorizes its matching private handoff without another permission question. Groups receive code-free acknowledgement. Existing callback state, expiry, account checks, and message retention remain required. No enforcement, settings, or storage changes are added.

Evidence

Synthetic policy checks covered 16 scenarios; 590 focused tests and seven revised prompt snapshots passed. Revised Telegram checks with a controlled service verified private delivery, group-to-private handoff, completed-status checks, rejected callbacks, and retained user messages. An earlier candidate disclosed a code in a group; that failure is retained. Revised runs still misstated code lifetime and one rejection explanation. This is guidance, not deterministic disclosure filtering, and proves the controlled service rather than third-party sign-in.

clawsweeper · 2026-09-05T15:52:08Z

🦞👀
ClawSweeper picked this up.

Pull request received. I will update this pull request when review starts.

ClawSweeper review complete

ClawSweeper finished reviewing this revision. The review result is being finalized.

View the workflow run.

clawsweeper · 2026-09-05T16:02:16Z

Codex review: passed. Reviewed September 7, 2026, 1:41 AM ET / 05:41 UTC.

ClawSweeper review

What this changes

Updates shared agent guidance, documentation, tests, and snapshots to allow private short-lived sign-in handoffs while keeping reusable secrets in protected entry.

Merge readiness

✅ Ready for maintainer review

This PR remains useful: current main and the latest release retain the blanket prohibition. The revised live observations address the earlier proof concern, and no blocking defect was found in the pinned patch.

Priority: P2
Reviewed head: d1e508a5aa88d1074e323480a851ca337970eeaa

Review scores

Measure	Result	What it means
Overall readiness	🐚 platinum hermit (4/6)	A focused shared-policy repair with relevant revised live observations and no actionable blocking finding.
Proof confidence	🐚 platinum hermit (4/6)	Sufficient (live_output): Captured revised Telegram Test Server observations exercise the shared prompt through an isolated production Gateway, real model, messaging delivery, and controlled login service: the requester receives the code privately, the group receives no code, and mismatched or cancelled callbacks fail without retry or false success. This addresses the prior authority-proof request for the tested boundary; it does not establish deterministic compliance across models or third-party vendors.
Patch quality	🐚 platinum hermit (4/6)	No actionable review findings were identified.

Verification

Check	Result	Evidence
Real behavior	Verified	Sufficient (live_output): Captured revised Telegram Test Server observations exercise the shared prompt through an isolated production Gateway, real model, messaging delivery, and controlled login service: the requester receives the code privately, the group receives no code, and mismatched or cancelled callbacks fail without retry or false success. This addresses the prior authority-proof request for the tested boundary; it does not establish deterministic compliance across models or third-party vendors.
Evidence reviewed	9 items	Pinned production change: The shared prompt requires selecting the requesting user's private conversation before code delivery, acknowledging a group handoff only after private delivery, and preserving state, PKCE, expiry, and account checks. Reusable secrets remain excluded from chat and tool arguments. Still necessary on main and release: Inspected current main and v2026.9.2; both retain the prohibition on authentication and pairing codes in transcripts. No verified merged replacement was established. Latest-release comparison: The supplied latest release still contains the original blanket credential guidance.
Findings	None	None.
Security	None	None.

How this fits together

OpenClaw assembles shared credential guidance into regular, Codex, Copilot, and setup-agent prompts. That guidance shapes how agents use existing messaging tools and service login flows.

flowchart TD
  A[User requests sign-in] --> B[Shared credential guidance]
  B --> C{Credential kind}
  C -->|Reusable secret| D[Protected entry]
  C -->|Short-lived code| E[Requester's private conversation]
  E --> F[Matching pending login flow]
  F --> G[Verified completion or rejection]

Before merge

None.

Agent review details

Security

None.

PR surface

Source +1, Tests -121, Docs +14. Total -106 across 14 files.

View PR surface stats

Area	Files	Added	Removed	Net
Source	3	7	6	+1
Tests	9	123	244	-121
Docs	2	16	2	+14
Config	0	0	0	0
Generated	0	0	0	0
Other	0	0	0	0
Total	14	146	252	-106

Review metrics

None.

Technical review

Best possible solution:

Keep one shared credential policy that permits requester-bound short-lived handoffs, preserves service checks, and reserves protected entry for reusable secrets.

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

No high-confidence current-main runtime reproduction was established: synthetic checks reportedly reproduced refusals, but the recorded real baseline completed the same private sign-in scenario.

Is this the best way to solve the issue?

Yes, the shared prompt is the narrow owner for correcting blanket refusal guidance; the patch preserves existing protected-entry and service-enforcement boundaries.

AGENTS.md: found and applied where relevant.

Codex review notes: model internal, reasoning medium; reviewed against f02c72916aeb.

Labels

Label changes:

add proof: sufficient: Contributor real behavior proof is sufficient. Captured revised Telegram Test Server observations exercise the shared prompt through an isolated production Gateway, real model, messaging delivery, and controlled login service: the requester receives the code privately, the group receives no code, and mismatched or cancelled callbacks fail without retry or false success. This addresses the prior authority-proof request for the tested boundary; it does not establish deterministic compliance across models or third-party vendors.
add rating: 🐚 platinum hermit: Overall readiness is 🐚 platinum hermit; proof is 🐚 platinum hermit and patch quality is 🐚 platinum hermit.
add status: 🚀 automerge armed: This PR is in ClawSweeper's automerge lane. Sufficient (live_output): Captured revised Telegram Test Server observations exercise the shared prompt through an isolated production Gateway, real model, messaging delivery, and controlled login service: the requester receives the code privately, the group receives no code, and mismatched or cancelled callbacks fail without retry or false success. This addresses the prior authority-proof request for the tested boundary; it does not establish deterministic compliance across models or third-party vendors.
add proof: telegram-e2e: This PR needs Telegram Test Server proof with the repository E2E skill. Private code delivery and code-free group acknowledgment are Telegram-visible behavior, covered by the reported revised Test Server observations.
remove rating: 🦪 silver shellfish: Current PR rating is rating: 🐚 platinum hermit, so this older rating label is no longer current.
remove merge-risk: 🚨 security-boundary: Current PR review selected no merge-risk labels.
remove status: 📣 needs proof: Current PR status label is status: 🚀 automerge armed.

Label justifications:

P2: This is a bounded correction to credential-flow guidance, with no demonstrated urgent runtime outage.
rating: 🐚 platinum hermit: Overall readiness is 🐚 platinum hermit; proof is 🐚 platinum hermit and patch quality is 🐚 platinum hermit.
status: 🚀 automerge armed: This PR is in ClawSweeper's automerge lane. Sufficient (live_output): Captured revised Telegram Test Server observations exercise the shared prompt through an isolated production Gateway, real model, messaging delivery, and controlled login service: the requester receives the code privately, the group receives no code, and mismatched or cancelled callbacks fail without retry or false success. This addresses the prior authority-proof request for the tested boundary; it does not establish deterministic compliance across models or third-party vendors.
proof: sufficient: Contributor real behavior proof is sufficient. Captured revised Telegram Test Server observations exercise the shared prompt through an isolated production Gateway, real model, messaging delivery, and controlled login service: the requester receives the code privately, the group receives no code, and mismatched or cancelled callbacks fail without retry or false success. This addresses the prior authority-proof request for the tested boundary; it does not establish deterministic compliance across models or third-party vendors.
proof: telegram-e2e: This PR needs Telegram Test Server proof with the repository E2E skill. Private code delivery and code-free group acknowledgment are Telegram-visible behavior, covered by the reported revised Test Server observations.

Evidence

What I checked:

Pinned production change: The shared prompt requires selecting the requesting user's private conversation before code delivery, acknowledging a group handoff only after private delivery, and preserving state, PKCE, expiry, and account checks. Reusable secrets remain excluded from chat and tool arguments. (src/agents/transcript-credential-safety.ts:7, d1e508a5aa88)
Still necessary on main and release: Inspected current main and v2026.9.2; both retain the prohibition on authentication and pairing codes in transcripts. No verified merged replacement was established. (src/agents/transcript-credential-safety.ts:7, f02c72916aeb)
Latest-release comparison: The supplied latest release still contains the original blanket credential guidance. (src/agents/transcript-credential-safety.ts:7, 3928bad9badf)
Revised production-flow observations: The complete captured PR body, context sourceRevision 04e6a17dfa273dd9bf60432dea04ff97312c3540b5adeaa56c64c4a25fb42efd, reports an isolated Gateway and real model on Telegram Test Server with a controlled login service. Revised observations include private code delivery, code-free group acknowledgment, completion-status verification, and mismatched-state/cancelled callback rejection without retry or replacement flow. These are real-flow observations, not merely the earlier synthetic model checks. The separate baseline also succeeded, and expiry/rejection narration inaccuracies remain explicitly outside the claimed fix. Raw local observations mentioned by the body were not present in the supplied scratch directory; this review uses the captured observations without claiming independent replay. (4d0c929f3cc2)
Proof remains applicable to final head: The final commit changes only four existing test files. The production prompt is unchanged from the head identified by the revised runtime evidence. Original-head parentage was also checked from its raw commit record. (d1e508a5aa88)
Codex dependency signal: The changed tests explicitly check native Codex developer instructions. The production builder includes the shared credential prompt, and thread-requests.ts passes that builder's result as developerInstructions, establishing the relevant dependency contract. (extensions/codex/src/app-server/thread-prompt.ts:141, d1e508a5aa88)

Likely related people:

Peter Steinberger: Suggested for follow-up; no historical authorship or introduction is verified. (role: unverified routing candidate; confidence: low)
Jason (Json): Suggested for follow-up; no historical authorship or introduction is verified. (role: unverified routing candidate; confidence: low)

Rating scale

Score	Internal tier	Crab rank	Meaning
6/6	S	🦀 challenger crab	Exceptional readiness
5/6	A	🦞 diamond lobster	Very strong readiness
4/6	B	🐚 platinum hermit	Good normal PR; ordinary maintainer review
3/6	C	🦐 gold shrimp	Useful, but confidence is limited
2/6	D	🦪 silver shellfish	Proof or implementation needs work
1/6	F	🧂 unranked krab	Not merge-ready
N/A	NA	🌊 off-meta tidepool	Rating does not apply

Overall follows the weaker of proof and patch quality.
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.

Workflow

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.

History

Review history (3 earlier review cycles)

reviewed 2026-09-05T16:01:27.486Z sha f9a66d4 :: needs real behavior proof before merge. :: none
reviewed 2026-09-07T05:01:46.847Z sha c219675 :: needs real behavior proof before merge. :: none
reviewed 2026-09-07T05:32:43.549Z sha d1e508a :: needs real behavior proof before merge. :: none

obviyus · 2026-09-07T02:34:21Z

@clawsweeper automerge

clawsweeper · 2026-09-07T02:34:24Z

🦞✅
ClawSweeper merged this PR after the passing review.

Source: clawsweeper[bot]
Feedback: structured ClawSweeper verdict: pass (sha=d1e508a5aa88d1074e323480a851ca337970eeaa)
Merge status: merged by ClawSweeper automerge
Merged at: 2026-09-07T05:43:27Z
Merge commit: 44dbc50d172e

What merged:

Updates shared agent guidance, documentation, tests, and snapshots to allow private short-lived sign-in handoffs while keeping reusable secrets in protected entry.

Automerge notes:

PR branch already contained follow-up commit before automerge: chore: refresh authorized credential flow prompt snapshots
PR branch already contained follow-up commit before automerge: fix: select private recipient before credential handoffs
PR branch already contained follow-up commit before automerge: test: align credential prompt delivery assertions

The automerge loop is complete.

Automerge progress:

2026-09-07 02:34:21 UTC review queued f9a66d4b4b24 (queued)

2026-09-07 02:36:03 UTC repair queued f9a66d4b4b24 (autonomous) Run: https://github.com/openclaw/clawsweeper/actions/runs/34076816706

2026-09-07 02:38:38 UTC repair started f9a66d4b4b24 (running) in 0s Run: https://github.com/openclaw/clawsweeper/actions/runs/34076816706 automerge-openclaw-openclaw-139196

2026-09-07 02:39:29 UTC validation plan f9a66d4b4b24 (passed) in 52s Run: https://github.com/openclaw/clawsweeper/actions/runs/34076816706 pnpm check:changed

2026-09-07 02:43:56 UTC Codex edit 1 1b917420d7ac (complete) in 5m 19s Run: https://github.com/openclaw/clawsweeper/actions/runs/34076816706 exit 0

2026-09-07 02:44:00 UTC validation and review 1 1b917420d7ac (running) in 5m 22s Run: https://github.com/openclaw/clawsweeper/actions/runs/34076816706 repair

2026-09-07 03:22:31 UTC repair finished f9a66d4b4b24 (blocked) in 43m 53s Run: https://github.com/openclaw/clawsweeper/actions/runs/34076816706 validation command failed (pnpm check:changed): $ node scripts/check-changed.mjs [check:changed] lanes=core, coreTests ... lowlisted. Coercion helper declarati...

2026-09-07 03:22:35 UTC repair completed f9a66d4b4b24 (no branch change) in 3s Run: https://github.com/openclaw/clawsweeper/actions/runs/34076816706 validation command failed (pnpm check:changed): $ node scripts/check-changed.mjs [check:changed] lanes=core, coreTests, extensionTests, docs, tooling [check:ch...

2026-09-07 05:42:15 UTC review passed d1e508a5aa88 (structured ClawSweeper verdict: pass (sha=d1e508a5aa88d1074e323480a851ca337970e...)

2026-09-07 05:43:29 UTC merged d1e508a5aa88 (merged by ClawSweeper automerge)

obviyus · 2026-09-07T05:11:14Z

landpr-approve-unrelated-red
pr: 139196
head: c219675
run: 34084921453 attempt: 1
failed: checks-node-changed-extensions-bundle-11 job=101627099195 conclusion=failure
check: openclaw/ci-gate

obviyus · 2026-09-07T05:45:48Z

Verified the ClawSweeper landing of d1e508a5aa88 as 44dbc50d172e, including the resulting tree.

Proof: final CI passed. Real-model Telegram Test Server checks passed private code delivery, confirmed completion, group-to-private handoff, mismatched/cancelled callback rejection, and user-message retention against a controlled service. The documented baseline and model-narration limitations remain unchanged.

Before policy	After policy
Blanket transcript prohibition on sign-in and pairing codes	Authorized short-lived handoffs to the requester's private conversation; reusable secrets remain protected

…claw#139196) Summary: - Updates shared agent guidance, documentation, tests, and snapshots to allow private short-lived sign-in handoffs while keeping reusable secrets in protected entry. Automerge notes: - PR branch already contained follow-up commit before automerge: chore: refresh authorized credential flow prompt snapshots - PR branch already contained follow-up commit before automerge: fix: select private recipient before credential handoffs - PR branch already contained follow-up commit before automerge: test: align credential prompt delivery assertions Validation: - ClawSweeper review passed for head d1e508a. - Required merge gates passed before the squash merge. Prepared head SHA: d1e508a Review: openclaw#139196 (comment) Co-authored-by: Ayaan Zaidi <hi@obviy.us> Approved-by: obviyus

…guidance Keep buildCredentialSafetyPrompt under its shipped SDK name for the compatibility window. Accept and ignore the deprecated string argument through 2026-11-30 while callers move to the controlToolsAvailable options object, with the window recorded in the compatibility registry and docs. Preserve the private login-code handoff from #139196. Add the terminal setup route only when control tools are unavailable, and carry that decision through core, Codex, and Copilot prompts. Keep the terminal line in one shared owner and regenerate the prompt fixtures. Restore the unchanged public surface counts of 4,447 exports and 2,629 callable exports without editing budgets or their tests.

@ejc3

…thorized logins (#143238) * fix(agents): drop the blanket credential prompt contract Blanket transcript credential guidance obstructed owner-authorized browser fills and workspace credential use. Embedded/local sessions also lacked the OpenClaw and Gateway tools needed to reach the suggested masked handoff. Remove the shared prompt builder from core, Codex, Copilot, the system agent, and the SDK. Keep egress guidance with the secrets tool description, add scoped masked-setup routing to setup tools, and direct sessions without control tools to masked terminal setup. Update browser login and SDK migration docs and regenerate prompt fixtures. Replace obsolete contract-text assertions with a capability-routing test that fails on the original missing terminal handoff. Preserve protected secret entry, transcript redaction, and browser tool behavior. Thanks @ejc3 for the reachability analysis in #128077 and @Stephane-fci for the workspace-credential report in #142010. Refs #120728 #120656 #128076 #128077 #142010 * fix(agents): keep the shipped credential prompt export with narrowed guidance Keep buildCredentialSafetyPrompt under its shipped SDK name for the compatibility window. Accept and ignore the deprecated string argument through 2026-11-30 while callers move to the controlToolsAvailable options object, with the window recorded in the compatibility registry and docs. Preserve the private login-code handoff from #139196. Add the terminal setup route only when control tools are unavailable, and carry that decision through core, Codex, and Copilot prompts. Keep the terminal line in one shared owner and regenerate the prompt fixtures. Restore the unchanged public surface counts of 4,447 exports and 2,629 callable exports without editing budgets or their tests. * test(agents): restore transport API after model endpoint tests The Arcee endpoint fixture loads a second transport graph through Jiti and replaces the shared worker's Responses test API. Restore its original property descriptor after the file so later schema contract tests use the Vite graph's configured host. Reproduced the three strict-flag failures in the CI shard and in an ordered polluter/victim pair. The repaired shard passes 8,495 tests; both file orders pass all nine tests and the standalone schema contract passes all four. Related: #143238

@ejc3

…thorized logins (openclaw#143238) * fix(agents): drop the blanket credential prompt contract Blanket transcript credential guidance obstructed owner-authorized browser fills and workspace credential use. Embedded/local sessions also lacked the OpenClaw and Gateway tools needed to reach the suggested masked handoff. Remove the shared prompt builder from core, Codex, Copilot, the system agent, and the SDK. Keep egress guidance with the secrets tool description, add scoped masked-setup routing to setup tools, and direct sessions without control tools to masked terminal setup. Update browser login and SDK migration docs and regenerate prompt fixtures. Replace obsolete contract-text assertions with a capability-routing test that fails on the original missing terminal handoff. Preserve protected secret entry, transcript redaction, and browser tool behavior. Thanks @ejc3 for the reachability analysis in openclaw#128077 and @Stephane-fci for the workspace-credential report in openclaw#142010. Refs openclaw#120728 openclaw#120656 openclaw#128076 openclaw#128077 openclaw#142010 * fix(agents): keep the shipped credential prompt export with narrowed guidance Keep buildCredentialSafetyPrompt under its shipped SDK name for the compatibility window. Accept and ignore the deprecated string argument through 2026-11-30 while callers move to the controlToolsAvailable options object, with the window recorded in the compatibility registry and docs. Preserve the private login-code handoff from openclaw#139196. Add the terminal setup route only when control tools are unavailable, and carry that decision through core, Codex, and Copilot prompts. Keep the terminal line in one shared owner and regenerate the prompt fixtures. Restore the unchanged public surface counts of 4,447 exports and 2,629 callable exports without editing budgets or their tests. * test(agents): restore transport API after model endpoint tests The Arcee endpoint fixture loads a second transport graph through Jiti and replaces the shared worker's Responses test API. Restore its original property descriptor after the file so later schema contract tests use the Vite graph's configured host. Reproduced the three strict-flag failures in the CI shard and in an ordered polluter/victim pair. The repaired shard passes 8,495 tests; both file orders pass all nine tests and the standalone schema contract passes all four. Related: openclaw#143238

…claw#139196) Summary: - Updates shared agent guidance, documentation, tests, and snapshots to allow private short-lived sign-in handoffs while keeping reusable secrets in protected entry. Automerge notes: - PR branch already contained follow-up commit before automerge: chore: refresh authorized credential flow prompt snapshots - PR branch already contained follow-up commit before automerge: fix: select private recipient before credential handoffs - PR branch already contained follow-up commit before automerge: test: align credential prompt delivery assertions Validation: - ClawSweeper review passed for head d1e508a. - Required merge gates passed before the squash merge. Prepared head SHA: d1e508a Review: openclaw#139196 (comment) Co-authored-by: Ayaan Zaidi <hi@obviy.us> Approved-by: obviyus (cherry picked from commit 44dbc50)

fix: allow authorized sign-in without blanket credential refusals

f9a66d4

obviyus marked this pull request as ready for review September 5, 2026 15:52

openclaw-barnacle Bot added docs Improvements or additions to documentation agents Agent runtime and tooling extensions: codex extensions: copilot size: M maintainer Maintainer-authored PR labels Sep 5, 2026

obviyus requested a review from steipete September 5, 2026 15:57

clawsweeper Bot added the clawsweeper:automerge Maintainer opted this PR into bounded ClawSweeper-reviewed automerge label Sep 7, 2026

chore: refresh authorized credential flow prompt snapshots

c219675

obviyus added 2 commits September 7, 2026 07:19

fix: select private recipient before credential handoffs

4d0c929

test: align credential prompt delivery assertions

d1e508a

clawsweeper Bot merged commit 44dbc50 into main Sep 7, 2026
187 checks passed

clawsweeper Bot deleted the codex/authorized-login-codes branch September 7, 2026 05:43

obviyus changed the title ~~fix: allow authorized credential flows without blanket refusals~~ feat(agents): allow authorized private credential handoffs Sep 7, 2026

clawsweeper Bot mentioned this pull request Sep 8, 2026

[Bug]: Codex runtime: secrets-store guidance makes the agent refuse the workspace's own credential files #142010

Closed

This was referenced Sep 9, 2026

fix: remove blanket masked-entry requirement for authorized logins #143208

Closed

fix(security): make transcript credential masking opt-in #128077

Closed

clawsweeper Bot mentioned this pull request Sep 9, 2026

fix(agents): drop the blanket credential prompt that blocked owner-authorized logins #143238

Merged

Navigation Menu

Repository navigation

feat(agents): allow authorized private credential handoffs - #139196

feat(agents): allow authorized private credential handoffs#139196
clawsweeper[bot] merged 4 commits into
mainfrom
codex/authorized-login-codes

obviyus commented Sep 5, 2026 •

edited

Loading

Uh oh!

clawsweeper Bot commented Sep 5, 2026 •

edited

Loading

Uh oh!

clawsweeper Bot commented Sep 5, 2026 •

edited

Loading

Security

PR surface

Review metrics

Technical review

Labels

Evidence

Rating scale

Workflow

History

Uh oh!

obviyus commented Sep 7, 2026

Uh oh!

clawsweeper Bot commented Sep 7, 2026 •

edited

Loading

Uh oh!

obviyus commented Sep 7, 2026

Uh oh!

Uh oh!

obviyus commented Sep 7, 2026 •

edited

Loading

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

1 participant

Conversation

obviyus commented Sep 5, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

What Problem This Solves

Why This Change Was Made

User Impact

Evidence

Uh oh!

clawsweeper Bot commented Sep 5, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

ClawSweeper review complete

Uh oh!

clawsweeper Bot commented Sep 5, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

ClawSweeper review

What this changes

Merge readiness

Review scores

Verification

How this fits together

Before merge

Security

PR surface

Review metrics

Technical review

Labels

Evidence

Rating scale

Workflow

History

Uh oh!

obviyus commented Sep 7, 2026

Uh oh!

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

Uh oh!

Uh oh!

obviyus commented Sep 7, 2026

Uh oh!

Uh oh!

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

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

1 participant

obviyus commented Sep 5, 2026 •

edited

Loading

clawsweeper Bot commented Sep 5, 2026 •

edited

Loading

clawsweeper Bot commented Sep 5, 2026 •

edited

Loading

clawsweeper Bot commented Sep 7, 2026 •

edited

Loading

obviyus commented Sep 7, 2026 •

edited

Loading