Redact tool output secrets by amknight · Pull Request #85196 · openclaw/openclaw

amknight · 2026-05-22T03:30:00Z

Summary

Expand tool-output redaction coverage for vendor token prefixes, credential fields, URL/query/form secrets, connection-string passwords, and obfuscated key separators.
Apply forced tool payload redaction to tool detail surfaces and share parsed redaction coverage with exec approval display bypass handling.
Extend sensitive URL query-name normalization and regression coverage for encoded, hidden, plus-spliced, and control-spliced keys.

Verification

pnpm format:fix src/logging/redact.ts src/logging/redact.test.ts src/shared/net/redact-sensitive-url.ts src/shared/net/redact-sensitive-url.test.ts src/infra/exec-approval-command-display.ts src/infra/exec-approval-command-display.test.ts
node scripts/run-vitest.mjs src/logging/redact.test.ts src/shared/net/redact-sensitive-url.test.ts src/infra/exec-approval-command-display.test.ts src/agents/pi-embedded-subscribe.tools.test.ts src/agents/tools/sessions-history-tool.test.ts src/gateway/ws-log.test.ts
git diff --check
AUTOREVIEW_AUTO_TESTS=0 .agents/skills/autoreview/scripts/autoreview --mode local

Behavior addressed: tool output and approval display text now redact a broader set of secret-like values by default, including vendor token prefixes, URL/query/form secrets, credential fields, and obfuscated sensitive keys.
Real environment tested: local Codex worktree on macOS with repository Vitest wrapper.
Exact steps or command run after this patch: ran focused redaction tests, touched-surface Vitest suite, diff whitespace check, formatter, and local autoreview.
Evidence after fix: touched-surface suite passed 10 test files and 214 tests; focused redaction suite passed 88 tests after replacing push-protection-triggering fixture literals with runtime-built strings.
Observed result after fix: secret-like values are masked across tool-detail redaction, URL helper redaction, and exec approval display bypass rendering without exposing the raw fixture values.
What was not tested: full repository test suite, build, and cross-platform Testbox lanes were not run.

clawsweeper · 2026-05-22T03:31:43Z

Codex review: needs maintainer review before merge. Reviewed June 10, 2026, 1:17 AM ET / 05:17 UTC.

Summary
The branch expands built-in redaction for tool/log/approval/URL surfaces and adds regression coverage for vendor token prefixes, credential fields, obfuscated keys, URL/query/form secrets, and base64 media false positives.

PR surface: Source +713, Tests +579. Total +1292 across 7 files.

Reproducibility: yes. source inspection gives a high-confidence path: current main's default prefilter lacks the new token/query/form triggers and tool detail redaction still depends on the older behavior, while the touched callers feed approval and tool display text through these helpers. I did not run the tests locally because this review is read-only and tests can create artifacts.

Review metrics: 1 noteworthy metric.

Default redaction policy: 1 built-in corpus expanded; 1 tool-detail path forced. Both affect existing users without a config change, so maintainers should intentionally accept the upgrade behavior before merge.

Merge readiness
Overall: 🐚 platinum hermit
Proof: 🦞 diamond lobster
Patch quality: 🐚 platinum hermit
Result: ready for maintainer review.

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

Rank-up moves:

Get maintainer/security acceptance for the broader default-redaction policy and the gateway-client follow-up scope.

Risk before merge

[P1] Existing users may see more log, diagnostic, URL, tool-detail, approval-display, and tool payload text masked after upgrade because the built-in default corpus is intentionally broader.
[P1] The patch changes a security boundary with complex regex, URL/form parsing, bitmap projection, and media-payload exceptions; focused tests and CI look coherent, but maintainer/security acceptance is still the merge gate.
[P1] An independent gateway-client URL redactor still has narrower query-name matching; the author called this out as a follow-up, so maintainers should decide whether that remaining gap is acceptable before treating URL-normalizer parity as complete.

Maintainer options:

Accept broader redaction and land (recommended)
A maintainer/security reviewer can merge once they intentionally accept the upgrade-visible masking changes and the separately tracked gateway-client cleanup scope.
Unify the sibling URL redactor first
If maintainers want URL-normalizer parity before merge, update the gateway-client redactor to reuse or match the net-policy normalization and add focused coverage.
Pause if default masking is too broad
If the expanded built-in corpus is not the desired default policy, pause this PR and ask for a narrower redaction set or an explicit rollout decision.

Next step before merge

[P2] Human maintainer/security review is the next action because this protected-label PR intentionally changes a security-boundary default and has no narrow automation repair remaining.

Security
Cleared: No concrete supply-chain or security regression was found in the diff; the remaining security concern is maintainer acceptance of the intentional redaction-boundary change.

Review details

Best possible solution:

Land after a maintainer/security reviewer accepts the broader default-redaction compatibility impact and the scoped gateway-client follow-up; otherwise fold that sibling URL-normalizer cleanup into this branch before merge.

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

Yes, source inspection gives a high-confidence path: current main's default prefilter lacks the new token/query/form triggers and tool detail redaction still depends on the older behavior, while the touched callers feed approval and tool display text through these helpers. I did not run the tests locally because this review is read-only and tests can create artifacts.

Is this the best way to solve the issue?

Yes, this is the best current fix shape for the central problem: it centralizes coverage in src/logging/redact.ts, reuses that coverage in exec approval display instead of duplicating a bitmap matcher, and adds focused regression tests. The one remaining sibling gap is the gateway-client redactor, which is acceptable only if maintainers agree to keep it as follow-up.

AGENTS.md: found and applied where relevant.

Codex review notes: model gpt-5.5, reasoning high; reviewed against e9671ed60327.

Label changes

Label justifications:

P2: The PR is a security hardening improvement with limited but real compatibility impact on displayed/logged text.
merge-risk: 🚨 compatibility: The diff intentionally broadens default masking and forces tool-detail redaction, so existing diagnostics and displays can change after upgrade.
merge-risk: 🚨 security-boundary: The changed regex/parsing/bitmap logic controls whether secrets are hidden or leaked across approval, tool, URL, and log surfaces.
rating: 🐚 platinum hermit: Overall readiness is 🐚 platinum hermit; proof is 🦞 diamond lobster and patch quality is 🐚 platinum hermit.
status: 👀 ready for maintainer look: ClawSweeper has no concrete contributor-facing blocker left for this PR. Not applicable: The external-contributor proof gate is not applicable because this is a maintainer-member PR, and the PR body plus follow-up comments provide terminal-style focused verification for the changed behavior.

Evidence reviewed

PR surface:

Source +713, Tests +579. Total +1292 across 7 files.

View PR surface stats

Area	Files	Added	Removed	Net
Source	3	785	72	+713
Tests	4	726	147	+579
Docs	0	0	0	0
Config	0	0	0	0
Generated	0	0	0	0
Other	0	0	0	0
Total	7	1511	219	+1292

What I checked:

Root policy read: Root AGENTS.md was read fully and its security-boundary, compatibility, whole-surface, proof, and protected-label review rules apply to this PR. (AGENTS.md:1, e9671ed60327)
Scoped policy check: The changed paths are under src, packages, infra, and logging; the only matching scoped guide read was src/agents/AGENTS.md, which is performance-focused and did not change the verdict. (src/agents/AGENTS.md:1, e9671ed60327)
Default redaction expansion: The PR head adds shared auth/form key sets, expanded vendor/token patterns, URL/userinfo/connection-string coverage, and a generated prefilter source list that is intended to stay in sync with default redaction patterns. (src/logging/redact.ts:137, d6428fcca660)
Approval display bypass handling: Exec approval display now computes raw and stripped redaction bitmaps from the shared logging redactor before union-masking bypassed positions, replacing the previous local bitmap implementation. (src/infra/exec-approval-command-display.ts:99, d6428fcca660)
Forced tool-detail redaction: redactToolDetail now routes through redactToolPayloadText, so tool display/detail surfaces keep tool-mode redaction even when general log redaction is off. (src/logging/redact.ts:994, d6428fcca660)
URL normalizer coverage: The net-policy URL redactor now normalizes decoded, plus-spliced, control, whitespace, and Hangul-filler-spliced query parameter names before classifying them as sensitive. (packages/net-policy/src/redact-sensitive-url.ts:13, d6428fcca660)

Likely related people:

steipete: Recent history and blame show Peter Steinberger carried net-policy extraction/refactors and logging/redaction fast-path maintenance around the touched URL and redaction surfaces. (role: recent area contributor; confidence: high; commits: f4c6c0aec49e, 1e7510ae103e, 2035f38ab244; files: packages/net-policy/src/redact-sensitive-url.ts, src/logging/redact.ts)
volcano303: Authored the merged Control UI tool payload redaction work that established forced tool payload redaction as the safety-boundary behavior this PR extends. (role: feature introducer; confidence: medium; commits: f3e8c50df392; files: src/logging/redact.ts)
drobison00: Authored the merged exec approval display sanitizer security hardening that this PR builds on for approval-prompt redaction and bypass rendering. (role: adjacent owner; confidence: medium; commits: 503b748a8ee1; files: src/infra/exec-approval-command-display.ts)
nimbleenigma: Recent blame shows follow-up work on persisted tool result detail redaction and resolveToolPayloadRedaction, adjacent to the forced tool-detail behavior changed here. (role: recent area contributor; confidence: medium; commits: 277eb16652dd; files: src/logging/redact.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.

clawsweeper · 2026-05-22T03:38:53Z

ClawSweeper PR egg

✨ Hatched: 🌱 uncommon Tiny Branchling

Hatch command

Comment @clawsweeper hatch when this PR is hatchable.

Hatchability rules:

Merged PRs are hatchable.
Open PRs are hatchable when they are status: 👀 ready for maintainer look, status: 🚀 automerge armed, or labeled clawsweeper:automerge.
Closed unmerged PRs are hatchable only when one of those hatchable labels is still present in the durable record.

Rarity: 🌱 uncommon.
Trait: finds missing screenshots.
Image traits: location release reef; accessory release bell; palette sunrise gold and clean white; mood bright-eyed; pose standing beside its cracked shell; shell paper lantern shell; lighting clean product lighting; background subtle branch markers.
Share on X: post this hatch
Copy: My PR egg hatched a 🌱 uncommon Tiny Branchling in ClawSweeper.

What is this egg doing here?

Eggs appear after the PR passes real-behavior proof. It is here for vibes, not verdicts: it does not change labels, ratings, merge decisions, or automation.
The shell reacts to review momentum: open follow-up work warms it up, re-review makes it wobble, and a clean final review lets it hatch.
Hatchability usually comes from sufficient real-behavior proof, no blocking P0/P1/P2 findings, no security attention needed, and clean correctness. A merged PR is already final, so merge makes the egg hatchable independently.
The hatch is seeded from this repository and PR number, so the same PR keeps the same creature; the reviewed head SHA can only change safe visual details.
Rarity is just collectible sparkle: 🥚 common, 🌱 uncommon, 💎 rare, ✨ glimmer, and 🌈 legendary.

- build DEFAULT_REDACT_PREFILTER_RE from sources covering every default pattern family: new vendor prefixes, webhook hosts, bare query/form keys, userinfo/connection-string passwords, and percent/plus/invisible obfuscated keys (including trailing separator splices) - run default-pattern redaction tests through the default options path and redact the vendor corpus per token so prefilter gaps fail tests - fix quoted standalone assignment values containing the other quote char or an unterminated quote; never re-mask *** placeholders - align net-policy URL query-name separator stripping with logging key normalization (Hangul fillers)

amknight · 2026-06-10T02:11:23Z

Addressed the ClawSweeper review in 038f346, plus two adjacent gaps found while re-reviewing.

Prefilter sync (the P1)
DEFAULT_REDACT_PREFILTER_RE is now assembled from DEFAULT_REDACT_PREFILTER_SOURCES, co-located with DEFAULT_REDACT_PATTERNS in src/logging/redact.ts, with triggers for every default this PR adds:

vendor prefixes and webhook hosts (gh[opsur]_, glpat-, gloas-, glsa_, sk_live_/sk_test_/rk_live_, SG., pypi-, do[opr]_v1_, dp.*, dckr_, AKIA/ASIA, api_org_, hooks.slack.com, discord, and the rest)
bare sensitive query/form keys (code=, jwt=, session=, credential, pass=)
URL userinfo and connection-string password shapes (://user:pass@)
structural triggers for obfuscated keys — percent escapes, plus splices, and invisible/control splices (including interior-plus-trailing separator combinations) — which literal key triggers can never cover.

Enforcement so this cannot drift again
The default-pattern suites in src/logging/redact.test.ts now exercise the real default-options path ({ mode: "tools" }, no explicit patterns), and the vendor corpus test redacts each token individually. A default pattern added without a prefilter trigger now fails its own test. The previous suites passed patterns: defaults, which skips the prefilter gate entirely — that is why the gap shipped silently. A parity test pins explicit-defaults ≡ default path.

Repro confirmed before the fix — the new tests against the prior head (4c820c0) fail exactly as the review predicted: the exec approval prompt renders deploy --with glpat-… unredacted, and ?code=…, jwt=…&safe=1, db pass=…, postgres://user:pw@…, and spliced client%5Fsecret=… all pass through redactSensitiveText untouched on the default path (7 failures). All pass after the fix.

Also fixed (found in re-review)

Quoted standalone assignment values containing the other quote character (password="it's-a-secret") or an unterminated quote leaked entirely: tagged main's single ([^\s&#]+) value class masked these, while the split quoted/unquoted patterns in this PR both failed to match. The quoted pattern now ends only at the matching close quote, the unquoted variant accepts one leading quote, and *** placeholders are never re-masked (which would have stripped their quote wrappers). Regression tests added.
packages/net-policy URL_QUERY_NAME_SEPARATOR_RE now also strips Hangul filler characters (category Lo, missed by \p{C}\p{Z}), matching FORM_BODY_KEY_SEPARATOR_RE in src/logging/redact.ts.

On the compatibility label: the forced redactToolDetail redaction matches the already-documented contract in https://docs.openclaw.ai/logging (Control UI tool-call events redact even when logging.redactSensitive is "off"), so that part closes a docs/code mismatch rather than adding a new contract.

Known follow-up (out of scope here): packages/gateway-client/src/client.ts keeps an independent substring-only isSensitiveUrlQueryParamName, so filler-spliced keys can still reach gateway-client error logs under a custom redactForLog. Tracked as a separate cleanup to reuse the net-policy normalizer.

Verification

node scripts/run-vitest.mjs src/logging/redact.test.ts packages/net-policy/src/redact-sensitive-url.test.ts src/infra/exec-approval-command-display.test.ts → 153 passed (105/16/32)
node scripts/run-vitest.mjs src/agents/embedded-agent-subscribe.tools.test.ts src/agents/tools/sessions-history-tool.test.ts src/gateway/ws-log.test.ts src/cli/gateway-cli/run.option-collisions.test.ts → 98 passed
node scripts/run-vitest.mjs src/agents/tool-display.test.ts src/agents/transcript-redact.test.ts src/logging/log-tail.test.ts src/infra/errors.test.ts → 86 passed
pnpm format:fix + scripts/run-oxlint.mjs on touched files (fixed one no-shadow in redactFormBodyLine); git diff --check clean
AUTOREVIEW_AUTO_TESTS=0 autoreview --mode local ran twice: round 1 flagged a trailing-separator prefilter bypass (client_seㅤcretㅤ=), fixed and regression-tested; round 2 clean — "no accepted/actionable findings".
Follow-up commit is net −104 lines (test conversions shrink more than the prefilter sources add).

@clawsweeper re-review

clawsweeper · 2026-06-10T02:11:26Z

🦞🧹
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/27248500276
Updated: 2026-06-10T02:21:23.168Z

- pure-base64-alphabet token prefixes (gAAAA, AKIA, ASIA, dapi, ATCTT3xFfG, ATATT, ATBB) now require a non-alphanumeric left boundary, skip explicit ;base64, payload spans, and run unchunked so chunk starts cannot fake the boundary or hide the container from the lookbehind - tokens after URL/path delimiters or assignments still mask; data-URL media survives redaction byte-identical (fixes chat media mirror CI) - regression tests: tiny-PNG data URL, in-blob plus boundary, chunk-aligned large data URL, reset-path Fernet token, path AWS key

amknight · 2026-06-10T05:08:49Z

CI failure root cause (checks-node-agentic-gateway-methods on 038f346: chat.directive-tags.test.ts > backs source reply media with an equivalent deduped delivery mirror) — fixed in d6428fc.

The failure was caused by this branch, in two parts:

This PR's Fernet pattern (gAAAA[A-Za-z0-9_=-]{20,}) had no left boundary, and tiny-PNG base64 contains gAAAA runs (zero-filled IHDR bytes). Redaction therefore mangled the test's data:image/png;base64,… media URL to …iVBORw0KGgoAAAANSUhEUgAAAAE…YII=, dropping the image block in the delivery mirror.
The prefilter sync in 038f346 is what made that latent over-redaction reachable on the default path — earlier CI runs passed only because the broken prefilter skipped the pattern. The new default-path enforcement surfaced a real over-redaction bug, which is exactly what it is for.

Fix: the seven new vendor prefixes whose alphabet is pure standard base64 (gAAAA, AKIA, ASIA, dapi, ATCTT3xFfG, ATATT, ATBB) now use a shared BASE64_SAFE_TOKEN_BOUNDARY:

a non-alphanumeric left boundary, so the prefix cannot fire mid-blob, while URL/path delimiters (/) and assignments (=) still qualify — https://app.example/reset/gAAAA… and /buckets/AKIA…/ keep redacting;
a negative lookbehind that skips explicit ;base64, payload spans, so a +- or /-adjacent prefix inside a data URL stays untouched;
these boundary patterns run unchunked: replacePatternBounded's 16 KiB chunking would otherwise fake a ^ boundary at chunk starts and hide the ;base64, container from the lookbehind on large media payloads.

Regression tests pin all of it: tiny-PNG data URL byte-identical through redaction, in-blob +gAAAA splice, a chunk-aligned ≥32 KiB data URL, plus positive cases for reset-path Fernet tokens and path-segment AWS keys.

Verification

node scripts/run-vitest.mjs src/gateway/server-methods/chat.directive-tags.test.ts → 212 passed (was 210/212 with the CI failure reproduced locally)
node scripts/run-vitest.mjs src/logging/redact.test.ts packages/net-policy/src/redact-sensitive-url.test.ts src/infra/exec-approval-command-display.test.ts → 156 passed
Consumer suites (embedded-agent-subscribe.tools, sessions-history-tool, ws-log, run.option-collisions, tool-display, transcript-redact, log-tail, errors) all green
AUTOREVIEW_AUTO_TESTS=0 autoreview --mode local iterated three more rounds: caught the /-delimiter over-narrowing and the chunk-boundary interaction (both fixed above); final round clean — "no actionable defects, patch is correct".

@clawsweeper re-review

clawsweeper · 2026-06-10T05:08:51Z

🦞🧹
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/27254679117
Updated: 2026-06-10T05:18:37.003Z

* redact tool output secrets * Expand tool-output secret redaction * fix(security): keep redaction prefilter in sync with expanded defaults - build DEFAULT_REDACT_PREFILTER_RE from sources covering every default pattern family: new vendor prefixes, webhook hosts, bare query/form keys, userinfo/connection-string passwords, and percent/plus/invisible obfuscated keys (including trailing separator splices) - run default-pattern redaction tests through the default options path and redact the vendor corpus per token so prefilter gaps fail tests - fix quoted standalone assignment values containing the other quote char or an unterminated quote; never re-mask *** placeholders - align net-policy URL query-name separator stripping with logging key normalization (Hangul fillers) * fix(security): keep base64-prefix redaction out of media payloads - pure-base64-alphabet token prefixes (gAAAA, AKIA, ASIA, dapi, ATCTT3xFfG, ATATT, ATBB) now require a non-alphanumeric left boundary, skip explicit ;base64, payload spans, and run unchunked so chunk starts cannot fake the boundary or hide the container from the lookbehind - tokens after URL/path delimiters or assignments still mask; data-URL media survives redaction byte-identical (fixes chat media mirror CI) - regression tests: tiny-PNG data URL, in-blob plus boundary, chunk-aligned large data URL, reset-path Fernet token, path AWS key --------- Co-authored-by: Alex Knight <[email protected]>

openclaw-barnacle Bot added size: XL maintainer Maintainer-authored PR labels May 22, 2026

amknight force-pushed the aknight/tool-output-secret-redaction branch from e1ddf1d to 292463b Compare May 22, 2026 03:45

openclaw-barnacle Bot added the cli CLI command changes label May 22, 2026

amknight self-assigned this May 22, 2026

amknight added 2 commits June 7, 2026 08:14

redact tool output secrets

7c6f21e

Expand tool-output secret redaction

4c820c0

amknight force-pushed the aknight/tool-output-secret-redaction branch from a6ae3b7 to 4c820c0 Compare June 7, 2026 15:23

clawsweeper Bot mentioned this pull request Jun 10, 2026

Feature Request: Masked Secrets - Prevent Agent from Accessing Raw API Keys #10659

Open

amknight merged commit 3faf669 into main Jun 10, 2026
159 checks passed

amknight deleted the aknight/tool-output-secret-redaction branch June 10, 2026 21:34

github-actions Bot mentioned this pull request Jun 10, 2026

📡 Upstream Digest — 2026-06-10 23:19 UTC curtismercier/openclaw-mods#1055

Open

This was referenced Jun 14, 2026

fix(acp): redact URL credentials in runtime errors #90211

Closed

[Feature]: Automatic Output Sanitization for Sensitive Data #43830

Open

Nas01010101 mentioned this pull request Jun 17, 2026

redact.ts over-masks ordinary aaaa-bbbb-cccc-dddd identifiers as Apple app-specific passwords, corrupting persisted transcript content #94254

Open

This was referenced Jun 30, 2026

fix(llm): preserve structured tool result text across providers #97742

Merged

Feature: logging.redactToolSensitive config option for tool payload redaction #98902

Open

fix(logging): redact Telegram bot tokens from timeout URLs #99428

Merged

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Uh oh!

Uh oh!

Redact tool output secrets#85196

Redact tool output secrets#85196
amknight merged 4 commits into
mainfrom
aknight/tool-output-secret-redaction

amknight commented May 22, 2026

Uh oh!

clawsweeper Bot commented May 22, 2026 •

edited

Loading

Uh oh!

clawsweeper Bot commented May 22, 2026 •

edited

Loading

Uh oh!

amknight commented Jun 10, 2026

Uh oh!

clawsweeper Bot commented Jun 10, 2026 •

edited

Loading

Uh oh!

amknight commented Jun 10, 2026

Uh oh!

clawsweeper Bot commented Jun 10, 2026 •

edited

Loading

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

1 participant

Uh oh!

Uh oh!

Conversation

amknight commented May 22, 2026

Uh oh!

clawsweeper Bot commented May 22, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

clawsweeper Bot commented May 22, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Hatch command

Uh oh!

amknight commented Jun 10, 2026

Uh oh!

clawsweeper Bot commented Jun 10, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

amknight commented Jun 10, 2026

Uh oh!

clawsweeper Bot commented Jun 10, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

1 participant

clawsweeper Bot commented May 22, 2026 •

edited

Loading

clawsweeper Bot commented May 22, 2026 •

edited

Loading

clawsweeper Bot commented Jun 10, 2026 •

edited

Loading

clawsweeper Bot commented Jun 10, 2026 •

edited

Loading