fix(memory, builtin backend): eagerly preload session transcript listeners at gateway startup by buyitsydney · Pull Request #76666 · openclaw/openclaw

buyitsydney · 2026-05-03T12:13:19Z

Summary (scope narrowed to builtin backend per Codex review)

Eagerly preload MemoryIndexManager at gateway startup for every agent whose agents.defaults.memorySearch.sources (or per-agent overlay) includes "sessions", when using the builtin session-file backend. This arms ensureSessionListener() before the first sessionTranscriptUpdate event fires, so /reset and /new archive emits actually land in the listener set and get indexed into chunks without the user having to run openclaw memory index --force.

Why it was broken

MemoryIndexManager is lazy-loaded on the first memory_search tool call. If a user types /reset or /new before that first search, the builtin archiveFileOnDisk emit lands in an empty SESSION_TRANSCRIPT_LISTENERS set and is silently dropped, so the resulting .jsonl.reset.<iso> / .jsonl.deleted.<iso> archives never enter chunks. Only memory index --force recovers them (which is #76311 / #76452's escape hatch but is not a cold-start fix).

Why builtin-only (Codex review response)

Thanks @clawsweeper for the P2 catch on the qmd path. The initial revision of this PR treated any successful getActiveMemorySearchManager result as listener-armed, but under the qmd backend that call returns the FallbackMemoryManager wrapper whose inner MemoryIndexManager is only constructed by fallbackFactory. So on qmd, merely obtaining the wrapper does NOT run ensureSessionListener(), and the cold-start race still drops sessionTranscriptUpdate emits.

Three options were on the table (per Codex):

Explicitly preload the builtin MemoryIndexManager listener owner even for qmd — requires reaching into the FallbackMemoryManager fallback slot.
Add qmd-native transcript event handling inside qmd-manager.ts.
Narrow this PR to builtin-only and defer qmd to a follow-up.

This revision takes option 3: added an explicit early-return guard if (params.cfg.memory?.backend === "qmd") { ... return; } at the top of startGatewayMemorySessionListeners, with a log line making the scope decision visible, plus a regression test (skips preload entirely when memory.backend is qmd (Codex review scope guard)). The changelog entry is also retitled Memory/sessions (builtin backend): and calls out the intentional qmd deferral.

Independent verification (builtin only)

CarHer builtin-backend deployment (carher-199 baseline). Three cold restarts, same container, identical /reset sequence:

	pre-preload	post-preload
archive emit lands in listener set	❌ dropped	✅ captured
`chunks` delta after 30s	+0	+41
verified commit	`main @ 9a49d143c819`	`HEAD of this PR`

What is NOT included

qmd backend session-listener arming (follow-up PR, tracked in body + changelog).
Changes to qmd-manager.ts transcript event handling.
Changes to FallbackMemoryManager fallback construction.

Refs #56131. AI-assisted.

clawsweeper · 2026-05-03T12:14:35Z

Codex review: found issues before merge. Reviewed July 3, 2026, 5:01 PM ET / 21:01 UTC.

Summary
The PR adds a gateway startup sidecar that preloads sessions-enabled builtin memory managers, adds focused gateway startup tests, and adds a CHANGELOG entry.

PR surface: Source +107, Tests +114, Docs +1. Total +222 across 4 files.

Reproducibility: yes. at source level. Current main does not construct a builtin MemoryIndexManager at gateway startup, and the current listener still returns before scheduling reset/deleted archive file events.

Review metrics: 2 noteworthy metrics.

Backend scope: 1 covered, 1 deferred. Builtin gets startup listener preloading while qmd remains out of scope, so maintainers should notice the asymmetric backend coverage before merge.
Startup lifetime: 1 cached manager per sessions-enabled builtin agent. The diff intentionally leaves non-transient managers open, which affects gateway resource and availability review before merge.

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:

Rebase against current main and pair or sequence this with fix(memory-core): keep live reindex of reset/deleted session archives #96132 or equivalent archive scheduling behavior.
Align the backend-scope comment with the builtin-only qmd skip.
Remove the release-owned CHANGELOG.md edit and keep release-note context in the PR body or squash message.

Risk before merge

[P1] Current main still returns before scheduling reset/deleted archive events in memory-core, so this branch can attach the listener but still miss the advertised live archive indexing path until fix(memory-core): keep live reindex of reset/deleted session archives #96132 or equivalent behavior is merged.
[P1] The new sidecar keeps a non-transient MemoryIndexManager alive for every sessions-enabled builtin agent, which is a gateway startup resource and lifecycle tradeoff maintainers should explicitly accept.
[P1] qmd remains intentionally out of scope, so merging this branch leaves qmd listener-owner parity to a follow-up.
[P1] A member comment notes the branch needs an update against current main for the required Dependency Guard check; that is a refresh requirement rather than a code defect.

Maintainer options:

Land archive scheduling first (recommended)
Update this branch on current main and land it only after fix(memory-core): keep live reindex of reset/deleted session archives #96132 or equivalent archive-event scheduling behavior is present.
Accept builtin-only manager lifetime
Maintainers can explicitly accept keeping builtin managers alive for sessions-enabled agents while leaving qmd listener parity to a separate follow-up.
Pause for a unified memory fix
If maintainers want one branch to cover listener attachment timing, archive scheduling, and qmd parity, pause or close this branch in favor of a consolidated fix.

Next step before merge

[P2] The remaining action is maintainer coordination with the open archive-scheduling PR, rebase/Dependency Guard refresh, and acceptance of builtin-only manager lifetime rather than a narrow automated repair.

Security
Cleared: No concrete security or supply-chain concern was found; the diff adds no dependencies, workflows, permissions, secrets handling, downloads, or external code execution.

Review findings

[P2] Pair preload with archive scheduling — src/gateway/server-startup-memory.ts:172-175
[P3] Align the backend-scope comment — src/gateway/server-startup-post-attach.ts:554-555
[P3] Remove the release-owned changelog edit — CHANGELOG.md:349

Review details

Best possible solution:

Rebase the branch, pair it with #96132 or equivalent archive scheduling, remove the changelog edit, align the qmd/builtin comment, and land only after maintainers accept the builtin-only manager lifetime tradeoff.

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

Yes, at source level. Current main does not construct a builtin MemoryIndexManager at gateway startup, and the current listener still returns before scheduling reset/deleted archive file events.

Is this the best way to solve the issue?

No, not as-is against current main. The builtin preload is the right timing fix, but it needs the archive-listener guard removed or already merged so the preloaded listener actually schedules reset/deleted archives.

Full review comments:

[P2] Pair preload with archive scheduling — src/gateway/server-startup-memory.ts:172-175
This call only arms the current memory listener. On current main, that listener returns before scheduling .jsonl.reset.* and .jsonl.deleted.* paths, so the advertised reset/new archive indexing fix still does not work unless this branch includes, waits for, or rebases after fix(memory-core): keep live reindex of reset/deleted session archives #96132 or equivalent behavior.
Confidence: 0.9
[P3] Align the backend-scope comment — src/gateway/server-startup-post-attach.ts:554-555
The comment says the sidecar runs for both qmd and builtin, but startGatewayMemorySessionListeners immediately returns for qmd. Please align the comment with the intentional builtin-only behavior so future readers do not rely on a qmd startup guarantee.
Confidence: 0.88
[P3] Remove the release-owned changelog edit — CHANGELOG.md:349
Repository policy keeps CHANGELOG.md release-owned for normal PRs. Please move the release-note context into the PR body or squash message and leave changelog generation to the release flow.
Confidence: 0.82

Overall correctness: patch is incorrect
Overall confidence: 0.87

AGENTS.md: found and applied where relevant.

Codex review notes: model internal, reasoning high; reviewed against 010b61746379.

Label changes

Label justifications:

P2: This is a normal-priority memory/session indexing bug fix with limited blast radius but real user impact for reset archive recall.
merge-risk: 🚨 session-state: The changed startup/listener path determines whether reset/deleted session archives become available to memory search during live sessions.
merge-risk: 🚨 availability: The diff adds a gateway startup sidecar that can keep background memory managers alive for every sessions-enabled builtin agent.
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 after-fix real-environment builtin-backend proof with repeated cold restarts and observed archive chunk deltas; current-main pairing still needs merge-context validation.
proof: sufficient: Contributor real behavior proof is sufficient. The PR body includes after-fix real-environment builtin-backend proof with repeated cold restarts and observed archive chunk deltas; current-main pairing still needs merge-context validation.

Evidence reviewed

PR surface:

Source +107, Tests +114, Docs +1. Total +222 across 4 files.

View PR surface stats

Area	Files	Added	Removed	Net
Source	2	107	0	+107
Tests	1	115	1	+114
Docs	1	1	0	+1
Config	0	0	0	0
Generated	0	0	0	0
Other	0	0	0	0
Total	4	223	1	+222

What I checked:

PR preload path: PR head adds startGatewayMemorySessionListeners and calls getActiveMemorySearchManager without a purpose for sessions-enabled non-qmd agents, leaving the manager open so the listener stays subscribed. (src/gateway/server-startup-memory.ts:172, b3f39fc38569)
Current main listener blocks archive scheduling: Current main returns immediately for archive artifact names before isSessionFileForAgent can call scheduleSessionDirty, while the later batch path still has explicit reset/deleted archive handling. (extensions/memory-core/src/memory/manager-sync-ops.ts:1462, 010b61746379)
Current main manager construction contract: An omitted purpose resolves to default, and non-transient MemoryIndexManager construction runs ensureSessionListener, so the preload mechanism can arm the listener when the builtin manager is constructed. (extensions/memory-core/src/memory/manager.ts:330, 010b61746379)
Archive emit is the live signal: archiveFileOnDisk emits sessionTranscriptUpdate for the archived path and documents that the event bus is the live channel because chokidar does not watch the sessions directory. (src/gateway/session-transcript-files.fs.ts:357, 010b61746379)
Related open scheduler fix: The related open PR explicitly removes the archive early return and includes before/after proof that reset/deleted archive emits reach dirty/sync scheduling; it is complementary to this preload branch, not a replacement. (extensions/memory-core/src/memory/manager-sync-ops.ts:1462, dc14d1a8d744)
Backend-scope comment mismatch: PR head comment says the sidecar runs for qmd and builtin, but startGatewayMemorySessionListeners returns immediately for qmd. (src/gateway/server-startup-post-attach.ts:554, b3f39fc38569)

Likely related people:

jalehman: The archive early return in the memory-core transcript listener appears in commit 4752524, which refactored the transcript update identity contract in the same listener path this PR depends on. (role: regression provenance and recent area contributor; confidence: high; commits: 475252453b4b; files: extensions/memory-core/src/memory/manager-sync-ops.ts, src/sessions/transcript-events.ts)
vincentkoc: Recent gateway memory startup work in commit 17be26b changed qmd startup manager lifetime behavior in the same startup files this PR extends. (role: gateway memory startup contributor; confidence: medium; commits: 17be26bc4f8b; files: src/gateway/server-startup-memory.ts, src/gateway/server-startup-post-attach.ts)
steipete: Commit 2b811fe added the lazy qmd gateway startup policy and scheduling structure that is the sibling startup surface for this builtin preload path. (role: gateway memory startup contributor; confidence: medium; commits: 2b811fe6d9e3; files: src/gateway/server-startup-memory.ts, src/gateway/server-startup-post-attach.ts)
buyitsydney: They authored prior merged archive visibility work and the earlier archive emit/delta handling commit, so they have context beyond merely opening this PR. (role: adjacent archive-indexing contributor; confidence: medium; commits: 2ffdb5d248a1, 04a8aa14cac1, 22b3e0d277c2; files: src/plugin-sdk/session-transcript-hit.ts, src/gateway/session-transcript-files.fs.ts, extensions/memory-core/src/memory/manager-sync-ops.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.

buyitsydney · 2026-05-03T13:39:54Z

Addressed @clawsweeper P2 finding in the latest push (c2cdae015b). Going with option 3: narrow scope to builtin backend.

Changes in this revision:

Added an explicit early-return guard in startGatewayMemorySessionListeners that skips preload entirely when params.cfg.memory?.backend === "qmd", with an info-level log line documenting the scope decision.
Added a regression test (skips preload entirely when memory.backend is qmd (Codex review scope guard)) asserting getActiveMemorySearchManager is never called under qmd.
Retitled PR + CHANGELOG entry to Memory/sessions (builtin backend) and explicitly documented qmd as out of scope with a follow-up PR deferred for arming the qmd listener owner.
Rebased onto latest upstream/main (past the 5.2 release and fix(memory): resolve archived transcript hits during session visibility filtering (follow-up to #76311, AI-assisted) #76452 merge).

Rationale: the qmd wrapper lazy-constructs its inner MemoryIndexManager via fallbackFactory, so a simple getActiveMemorySearchManager call on the qmd path does not arm the listener. That's a separate design problem in the FallbackMemoryManager / qmd-manager path that deserves its own PR and review, not a bolt-on inside this one.

Ready for re-review.

buyitsydney · 2026-05-03T14:03:00Z

P3 addressed in 1da8e474ca:

Added experimental: { sessionMemory: true } to createBuiltinSessionsConfig fixture so resolveMemorySearchConfig keeps the sessions source and the preload path actually reaches getActiveMemorySearchManager.
Strengthened the close-check regression test with expect(getMemorySearchManagerMock).toHaveBeenCalledWith(expect.objectContaining({ agentId: "main" })) — so any future regression that silently drops sessions-source resolution (or skips manager acquisition) fails the test instead of passing vacuously.
Also added the flag to createBuiltinNoSessionsConfig for symmetry; that fixture's intent is 'the agent has no sessions source', and the flag not being the gating reason makes that clearer.

Ready for re-review.

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/25327143305
Updated: 2026-05-04T15:23:57.770Z

buyitsydney · 2026-05-03T15:08:49Z

FYI on the red check: the 3 failing jobs are all the same upstream/main flake, not from this PR's diff.

Failed test: src/gateway/server-startup-web-fetch-bind.test.ts > binds HTTP with credential-free tools.web.fetch config without fetch provider discovery

Error chain:

[vitest] There was an error when mocking a module. 
Caused by: ReferenceError: Cannot access '__vi_import_1__' before initialization
 ❯ src/gateway/test-helpers.mocks.ts:204:3
 ❯ src/agents/context.ts:6:1
 ❯ src/gateway/session-utils.ts:12:1

This is a vitest vi.mock hoisting / circular-import race, not a test assertion failure. None of those files are in this PR's diff (only server-startup-memory.ts/.test.ts, server-startup-post-attach.ts, CHANGELOG.md). The previous upstream commit literally titled ci: retrigger checks (lint + dist-shard flakes in previous run) confirms this is a known-flaky shard.

Just rebased onto the latest upstream/main (past 9e56cfc + friends) to retrigger CI; if the flake persists after re-run, it's a maintainer side fix, not this PR.

My 11 tests in server-startup-memory.test.ts all pass (✓ 11 tests).

buyitsydney · 2026-05-04T08:12:18Z

Friendly ping — CI is now clean on the latest head (05638b1fa57): 100/100 check runs completed, 79 success / 20 skipped / 1 neutral, 0 failures.

This PR is intentionally scoped to the builtin backend startup listener preload; it leaves the qmd follow-up out of scope per review feedback. Ready for maintainer review when you have bandwidth. Thanks!

…startup MemoryIndexManager subscribes to sessionTranscriptUpdate events only when it is instantiated, via ensureSessionListener() inside its constructor. Before this change the manager was lazy-loaded on the first memory_search tool call, which means any sessionTranscriptUpdate emit landing earlier in the process lifetime went to an empty SESSION_TRANSCRIPT_LISTENERS set and was silently lost. The most user-visible casualty was the archiveFileOnDisk emit triggered by /reset and /new: the resulting .jsonl.reset.<iso> / .jsonl.deleted.<iso> archive file stayed on disk but never entered the chunks table, so memory_search could not surface it until the user manually ran `openclaw memory index --force`. The existing sidecars.memory startup task (startGatewayMemoryBackend) only runs for the qmd backend (via resolveGatewayMemoryStartupPolicy returning mode=off on non-qmd), so builtin-backend deployments had no eager-preload path at all. Fix --- Add a sibling startup task sidecars.memory-session-listeners that runs for every backend. It calls getActiveMemorySearchManager without a purpose (so MemoryIndexManager.get picks the non-transient 'default' branch and ensureSessionListener() runs), then deliberately leaves the manager cached rather than closing it, because close() would unsubscribe the listener we just attached. Per-agent failures are swallowed with log.warn; the task as a whole is wrapped in setImmediate and never blocks gateway boot. Verification ------------ Independently verified on a builtin-backend deployment running a single default agent with memorySearch.sources = ["memory", "sessions"]: - Before this patch, cold-start (container recreated, no prior memory_search call): /reset archive file generated on disk, but chunks delta = 0 and files delta = 0 after 3 min of waiting. - After this patch, identical cold-start: /reset archive indexed within ~30 s (chunks delta = 41, files +1, new path carries the .jsonl.reset.<iso> suffix). Repeated across three consecutive container recreates with stable results. Refs openclaw#56131. AI-assisted.

buyitsydney · 2026-05-06T05:01:42Z

Gentle ping @steipete @vincentkoc — this PR is scoped to the builtin-backend startup listener preload and has been sitting ready for ~45h.

Status:

,
CI: 100/100 check runs — 79 success / 20 skipped / 1 neutral, 0 failures on head b3f39fc385
clawsweeper verdict: needs-human, confidence=high — no repair job needed, normal maintainer review/merge is the remaining action
qmd listener arming intentionally left out of scope as you suggested (separate follow-up)

Happy to rebase if you want it on latest main (currently behind 498 but still merges clean). Appreciate your time!

openclaw-barnacle Bot added gateway Gateway runtime size: S labels May 3, 2026

buyitsydney force-pushed the fix/memory-session-listeners-eager-preload branch from 6ccc895 to c2cdae0 Compare May 3, 2026 13:39

openclaw-barnacle Bot added size: M and removed size: S labels May 3, 2026

buyitsydney changed the title ~~fix(memory): eagerly preload session transcript listeners at gateway startup~~ fix(memory, builtin backend): eagerly preload session transcript listeners at gateway startup May 3, 2026

buyitsydney force-pushed the fix/memory-session-listeners-eager-preload branch from c2cdae0 to 1da8e47 Compare May 3, 2026 14:02

buyitsydney marked this pull request as ready for review May 3, 2026 14:03

buyitsydney force-pushed the fix/memory-session-listeners-eager-preload branch from c5615a8 to bbeae3e Compare May 3, 2026 14:49

buyitsydney force-pushed the fix/memory-session-listeners-eager-preload branch from bbeae3e to 05638b1 Compare May 3, 2026 15:08

buyitsydney mentioned this pull request May 4, 2026

[memory] sessions source counter overflow: tombstones and reset snapshots indexed as live files #77338

Closed

buyitsydney added 2 commits May 4, 2026 23:12

ci: retrigger checks (lint + dist-shard flakes in previous run)

b3f39fc

buyitsydney force-pushed the fix/memory-session-listeners-eager-preload branch from 05638b1 to b3f39fc Compare May 4, 2026 15:14

clawsweeper Bot added the proof: sufficient ClawSweeper judged the real behavior proof convincing. label May 14, 2026

openclaw-barnacle Bot added triage: needs-real-behavior-proof Candidate: external PR needs after-fix proof from a real setup. and removed proof: sufficient ClawSweeper judged the real behavior proof convincing. labels May 14, 2026

clawsweeper Bot added the proof: sufficient ClawSweeper judged the real behavior proof convincing. label May 14, 2026

openclaw-barnacle Bot removed the proof: sufficient ClawSweeper judged the real behavior proof convincing. label May 14, 2026

clawsweeper Bot added the proof: sufficient ClawSweeper judged the real behavior proof convincing. label May 14, 2026

openclaw-barnacle Bot removed the proof: sufficient ClawSweeper judged the real behavior proof convincing. label May 14, 2026

clawsweeper Bot added the proof: sufficient ClawSweeper judged the real behavior proof convincing. label May 15, 2026

openclaw-barnacle Bot removed the proof: sufficient ClawSweeper judged the real behavior proof convincing. label May 15, 2026

clawsweeper Bot added the proof: sufficient ClawSweeper judged the real behavior proof convincing. label May 15, 2026

clawsweeper Bot added rating: 🌊 off-meta tidepool PR readiness rating does not apply to this item. and removed status: 👀 ready for maintainer look ClawSweeper has no concrete contributor-facing blocker left for this PR. labels Jun 14, 2026

openclaw-barnacle Bot added the triage: needs-real-behavior-proof Candidate: external PR needs after-fix proof from a real setup. label Jun 14, 2026

openclaw-barnacle Bot removed the triage: needs-real-behavior-proof Candidate: external PR needs after-fix proof from a real setup. label Jun 14, 2026

openclaw-barnacle Bot added the triage: needs-pr-context Candidate: external PR body lacks required problem context or evidence. label Jun 22, 2026

yetval mentioned this pull request Jun 23, 2026

fix(memory-core): keep live reindex of reset/deleted session archives #96132

Merged

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Uh oh!

Uh oh!

fix(memory, builtin backend): eagerly preload session transcript listeners at gateway startup#76666

fix(memory, builtin backend): eagerly preload session transcript listeners at gateway startup#76666
buyitsydney wants to merge 2 commits into
openclaw:mainfrom
buyitsydney:fix/memory-session-listeners-eager-preload

buyitsydney commented May 3, 2026 •

edited

Loading

Uh oh!

clawsweeper Bot commented May 3, 2026 •

edited

Loading

Uh oh!

buyitsydney commented May 3, 2026

Uh oh!

buyitsydney commented May 3, 2026 •

edited by clawsweeper Bot

Loading

Uh oh!

buyitsydney commented May 3, 2026

Uh oh!

buyitsydney commented May 4, 2026

Uh oh!

buyitsydney commented May 6, 2026

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

Uh oh!

Uh oh!

Conversation

buyitsydney commented May 3, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Summary (scope narrowed to builtin backend per Codex review)

Why it was broken

Why builtin-only (Codex review response)

Independent verification (builtin only)

What is NOT included

Uh oh!

clawsweeper Bot commented May 3, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

buyitsydney commented May 3, 2026

Uh oh!

buyitsydney commented May 3, 2026 • edited by clawsweeper Bot Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

buyitsydney commented May 3, 2026

Uh oh!

buyitsydney commented May 4, 2026

Uh oh!

buyitsydney commented May 6, 2026

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

buyitsydney commented May 3, 2026 •

edited

Loading

clawsweeper Bot commented May 3, 2026 •

edited

Loading

buyitsydney commented May 3, 2026 •

edited by clawsweeper Bot

Loading