fix(lmstudio): bound model load success response body to prevent OOM by Alix-007 · Pull Request #96042 · openclaw/openclaw

Alix-007 · 2026-06-23T08:31:16Z

What Problem This Solves

ensureLmstudioModelLoaded in extensions/lmstudio/src/models.fetch.ts POSTs to LM Studio's /api/v1/models/load and reads the success response with an unbounded await response.json(). LM Studio is a self-hosted/local server endpoint that the agent reaches over HTTP; a misbehaving, misconfigured, or compromised server (or anything sitting on the configured baseUrl, including an SSRF-redirected target) can return a success body with no Content-Length and stream an arbitrarily large JSON payload.

Because response.json() buffers the entire body before parsing, that unbounded body is fully read into memory on the provider/discovery path before any size check runs. That is a memory-pressure / hang vector on a code path that runs during model discovery and before first inference — exactly the kind of unbounded external-response read the response-limit campaign has been closing across the codebase.

The sibling error body on the same endpoint was already bounded (readResponseTextLimited, 8 KiB) in a prior change. This PR closes the remaining unbounded read: the success JSON body.

Changes

extensions/lmstudio/src/models.fetch.ts: replace the unbounded await response.json() on the /api/v1/models/load success path with the shared, already-bounded readProviderJsonResponse<LmstudioLoadResponse>(...) helper (re-exported from openclaw/plugin-sdk/provider-http, same helper the /api/v1/models discovery path already uses). It reads the body through readResponseWithLimit under the 16 MiB provider-JSON cap, cancels the stream on overflow, and wraps malformed JSON with the caller's label. No new abstraction is introduced and the redundant local try/catch is removed.
extensions/lmstudio/src/models.test.ts: migrate the model fetch/load mocks from bare { ok, json } placeholders to real Response objects (the bounded readers require a real body stream), and add a regression test that streams an oversized (>16 MiB) success body and asserts the load path throws a bounded error and cancels the stream instead of buffering it all. Updates the malformed-load assertion to the helper's owned message.

Real behavior proof

Behavior addressed: unbounded buffering of the LM Studio /api/v1/models/load success JSON body; after the fix the read is capped at 16 MiB and the stream is cancelled on overflow.
Real environment tested: a local node:http server (real sockets, chunked transfer-encoding, no Content-Length) driving the real exported ensureLmstudioModelLoaded / readProviderJsonResponse via the global fetch (real Response body streams). Run with node --import tsx proof.mts.
Exact steps: (1) normal small { "status": "loaded" } response, end-to-end ensureLmstudioModelLoaded; (2) oversized 20 MiB chunked success body, end-to-end; (3) a body that hangs forever after a prefix, driving readProviderJsonResponse with an explicit small cap; (4) negative control reading the same 20 MiB body with the pre-fix unbounded await response.text().
Evidence after fix: (1) returns "proof-model" (still parses normally); (2) throws LM Studio model load: JSON response exceeds 16777216 bytes and the server observed only ~17.5 MiB sent before the socket closed (vs 20 MiB total) — the stream was cancelled mid-flight; (3) throws ... exceeds 2048 bytes and the server socket was closed (cancelled) instead of hanging; (4) the unbounded read consumed the full 20 MiB, confirming the cap is load-bearing.
Observed result: oversized/hanging bodies are stopped at the cap with a bounded error and the upstream stream is cancelled; well-formed small bodies still parse.
What was not tested: a real third-party LM Studio binary returning a pathological body (used a faithful local HTTP server instead); behavior under genuine OS memory exhaustion (the goal is to prevent reaching it).

Evidence

[proof] 本地 server 启动 http://127.0.0.1:36821
PASS  normal-small 端到端解析不破 — ensureLmstudioModelLoaded 返回 proof-model
PASS  oversized 抛 bounded error — error="LM Studio model load: JSON response exceeds 16777216 bytes"
PASS  oversized server 发送量 < 全量 20MiB(stream 被提前 cancel) — server 实发 17.50MiB (cap 16MiB,远小于 20MiB 全量)
PASS  hang body 到 cap 即抛 bounded error(不无限挂起) — error="LM Studio model load: JSON response exceeds 2048 bytes"
PASS  hang body cap 命中后 server 端 socket 被 cancel/close — server 实发 720922 bytes 后 socket close=true
PASS  negative control:无界 await .text() 全量读入 20MiB(证明 cap load-bearing) — 无界读到 20.00MiB 文本 / server 实发 20.00MiB(对照:bounded 仅读到 cap)
[proof] ALL PASS

Unit tests: node scripts/run-vitest.mjs extensions/lmstudio/src/models.test.ts --run → 24 passed (24), including the new bounds oversized model load success bodies. oxlint and tsgo (extensions source + test) clean on the changed files.

Label: security

This is the symmetric companion to the response-limit campaign (e.g. #95103 / #95108 / #95218): the LM Studio discovery /api/v1/models path and the /api/v1/models/load error body were already bounded; this PR finishes the set by bounding the /api/v1/models/load success body with the same shared helper and cap.

AI-assisted: implemented and proof-tested with AI assistance; reviewed by a human before submission.

The /api/v1/models/load success path read the response with an unbounded await response.json(), so a misbehaving or compromised LM Studio server could stream an arbitrarily large JSON body that is fully buffered into memory before any size check. Read it through the shared byte-capped readProviderJsonResponse helper instead (16 MiB provider-JSON cap, cancels the stream on overflow, wraps malformed JSON), matching the discovery path and the already-bounded error body. Migrate the model fetch/load test mocks to real Response objects (the bounded readers need a real body stream) and add a regression test that streams an oversized success body and asserts a bounded error plus stream cancellation. Label: security

Alix-007 · 2026-06-23T10:19:12Z

@clawsweeper review

clawsweeper · 2026-06-23T10:19:15Z

🦞🧹
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.

clawsweeper · 2026-06-23T21:05:55Z

Codex review: needs maintainer review before merge. Reviewed June 23, 2026, 5:09 PM ET / 21:09 UTC.

Summary
The PR replaces LM Studio model-load success response.json() with the shared bounded provider JSON reader and updates tests to use real Response bodies plus an oversized-stream cancellation regression.

PR surface: Source +2, Tests +27. Total +29 across 2 files.

Reproducibility: yes. Source inspection shows current main still buffers the LM Studio model-load success body with response.json(), and the PR body provides a real-socket harness with a negative control for oversized stream behavior; I did not rerun it in this read-only review.

Review metrics: none identified.

Stored data model
Persistent data-model change detected: serialized state: extensions/lmstudio/src/models.test.ts. Confirm migration or upgrade compatibility proof before merge.

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

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

Next step before merge

[P2] No repair lane is needed; this contributor PR is focused, proof-backed, and has no actionable findings from this review.

Security
Cleared: Cleared: the diff only routes an existing external response body through an established byte-capped reader and adds tests, with no new dependencies, workflows, package scripts, permissions, or secret handling.

Review details

Best possible solution:

Land the narrow helper reuse after normal maintainer checks so LM Studio load success bodies use the same 16 MiB provider JSON cap as other provider JSON reads.

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

Yes. Source inspection shows current main still buffers the LM Studio model-load success body with response.json(), and the PR body provides a real-socket harness with a negative control for oversized stream behavior; I did not rerun it in this read-only review.

Is this the best way to solve the issue?

Yes. Reusing the existing readProviderJsonResponse helper is the narrowest maintainable fix because the helper already owns the 16 MiB provider JSON cap, overflow error, malformed JSON label, and stream cancellation semantics.

AGENTS.md: found and applied where relevant.

Codex review notes: model internal, reasoning high; reviewed against 654544b6b7c4.

Label changes

Label changes:

add P2: This is a focused provider-plugin security/availability hardening for an OOM-capable response body path with limited blast radius.
add proof: sufficient: Contributor real behavior proof is sufficient. The PR body includes terminal output from a local real-socket HTTP harness showing normal parsing, oversized success-body rejection, hanging-body cancellation at a cap, and a pre-fix unbounded negative control.
add rating: 🦞 diamond lobster: Overall readiness is 🦞 diamond lobster; proof is 🦞 diamond lobster and patch quality is 🦞 diamond lobster.
add status: 👀 ready for maintainer look: ClawSweeper has no concrete contributor-facing blocker left for this PR. Sufficient (terminal): The PR body includes terminal output from a local real-socket HTTP harness showing normal parsing, oversized success-body rejection, hanging-body cancellation at a cap, and a pre-fix unbounded negative control.

Label justifications:

P2: This is a focused provider-plugin security/availability hardening for an OOM-capable response body path with limited blast radius.
rating: 🦞 diamond lobster: Overall readiness is 🦞 diamond lobster; proof is 🦞 diamond lobster and patch quality is 🦞 diamond lobster.
status: 👀 ready for maintainer look: ClawSweeper has no concrete contributor-facing blocker left for this PR. Sufficient (terminal): The PR body includes terminal output from a local real-socket HTTP harness showing normal parsing, oversized success-body rejection, hanging-body cancellation at a cap, and a pre-fix unbounded negative control.
proof: sufficient: Contributor real behavior proof is sufficient. The PR body includes terminal output from a local real-socket HTTP harness showing normal parsing, oversized success-body rejection, hanging-body cancellation at a cap, and a pre-fix unbounded negative control.

Evidence reviewed

PR surface:

Source +2, Tests +27. Total +29 across 2 files.

View PR surface stats

Area	Files	Added	Removed	Net
Source	1	8	6	+2
Tests	1	104	77	+27
Docs	0	0	0	0
Config	0	0	0	0
Generated	0	0	0	0
Other	0	0	0	0
Total	2	112	83	+29

What I checked:

Current main still has the unbounded success read: On current main, ensureLmstudioModelLoaded reads the successful /api/v1/models/load response with await response.json() after the non-OK body path is already bounded. (extensions/lmstudio/src/models.fetch.ts:290, 654544b6b7c4)
PR changes the load success path to the shared bounded helper: The PR branch imports readProviderJsonResponse and uses it for LM Studio model load, keeping the existing status validation after parsing. (extensions/lmstudio/src/models.fetch.ts:292, 90c3b9653e10)
Shared helper contract is already bounded on main: readProviderJsonResponse uses the 16 MiB provider JSON cap and readResponseWithLimit, wrapping overflow as <label>: JSON response exceeds <n> bytes and malformed JSON as <label>: malformed JSON response. (src/agents/provider-http-errors.ts:297, 654544b6b7c4)
Bounded reader cancels overflow streams: readResponseWithLimit truncates when the next chunk exceeds the byte cap and calls reader.cancel() before returning an overflow error. (packages/media-core/src/read-response-with-limit.ts:96, 654544b6b7c4)
PR adds focused regression coverage: The new test streams more than the provider JSON cap from the model-load success response, expects a bounded JSON-size error, verifies stream cancellation, and verifies the full 32 MiB body was not read. (extensions/lmstudio/src/models.test.ts:589, 90c3b9653e10)
Related helper provenance: The generic readProviderJsonResponse byte-cap helper came from merged pull request fix(agents): bound provider JSON response reads #95218, whose merge commit is present on current main and is the right shared primitive for this LM Studio endpoint. (src/agents/provider-http-errors.ts:297, 2592f8a51a4e)

Likely related people:

vincentkoc: The current checkout history shows Vincent Koc introducing the present LM Studio model fetch/load file and tests, including the unbounded load success read that this PR changes. (role: recent area contributor; confidence: high; commits: fa263affd584; files: extensions/lmstudio/src/models.fetch.ts, extensions/lmstudio/src/models.test.ts)
Alix-007: Alix-007 authored the merged provider JSON response cap in fix(agents): bound provider JSON response reads #95218, which supplies the shared helper this PR now applies to LM Studio. (role: recent shared-helper contributor; confidence: high; commits: 2592f8a51a4e; files: src/agents/provider-http-errors.ts, src/agents/provider-http-errors.test.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.

sallyom · 2026-06-24T13:02:45Z

LGTM/merge-ready:

Local autoreview in the PR worktree against origin/main: clean, no accepted/actionable findings. I also reviewed upgrade/compatibility risk and found no blocking concerns: no config, persisted state, migration, plugin SDK, or LM Studio API compatibility blocker. The only user-visible behavior change is the intended bounded-success-body failure path/error text for malformed or oversized JSON.
CI/check rollup is green, and focused local proof passed: node scripts/run-vitest.mjs extensions/lmstudio/src/ models.test.ts --run.
Best-fix verdict: this is the right narrow fix because it reuses the existing shared bounded provider JSON reader instead of adding LM Studio-local parsing.

…penclaw#96042) The /api/v1/models/load success path read the response with an unbounded await response.json(), so a misbehaving or compromised LM Studio server could stream an arbitrarily large JSON body that is fully buffered into memory before any size check. Read it through the shared byte-capped readProviderJsonResponse helper instead (16 MiB provider-JSON cap, cancels the stream on overflow, wraps malformed JSON), matching the discovery path and the already-bounded error body. Migrate the model fetch/load test mocks to real Response objects (the bounded readers need a real body stream) and add a regression test that streams an oversized success body and asserts a bounded error plus stream cancellation. Label: security

…sonResponse Replace the local readGoogleChatJsonResponse and readGoogleChatCertsResponse wrappers with the existing SDK helper readProviderJsonResponse (from openclaw/plugin-sdk/provider-http) so the Google Chat API JSON responses are bounded at 16 MiB, matching the non-streaming cap already used by 15+ other extensions. What changed: - extensions/googlechat/src/api.ts: readGoogleChatJsonResponse now delegates to readProviderJsonResponse. Removed the local try/catch wrapper. - extensions/googlechat/src/auth.ts: readGoogleChatCertsResponse now delegates to readProviderJsonResponse. Error message preserved. Removed the local try/catch wrapper. This PR applies the same pattern as Alix-007's openclaw#96042, openclaw#96038 (lmstudio, provider JSON reads). No SDK promotion needed — readProviderJsonResponse is already available in openclaw/plugin-sdk/provider-http.

…sonResponse (#96772) * fix(googlechat): replace unbounded response.json() with readProviderJsonResponse Replace the local readGoogleChatJsonResponse and readGoogleChatCertsResponse wrappers with the existing SDK helper readProviderJsonResponse (from openclaw/plugin-sdk/provider-http) so the Google Chat API JSON responses are bounded at 16 MiB, matching the non-streaming cap already used by 15+ other extensions. What changed: - extensions/googlechat/src/api.ts: readGoogleChatJsonResponse now delegates to readProviderJsonResponse. Removed the local try/catch wrapper. - extensions/googlechat/src/auth.ts: readGoogleChatCertsResponse now delegates to readProviderJsonResponse. Error message preserved. Removed the local try/catch wrapper. This PR applies the same pattern as Alix-007's #96042, #96038 (lmstudio, provider JSON reads). No SDK promotion needed — readProviderJsonResponse is already available in openclaw/plugin-sdk/provider-http. * fix(googlechat): add inline bounded-read regression tests Co-Authored-By: Claude <[email protected]> * fix(googlechat): remove unused variable flagged by oxlint Co-Authored-By: Claude <[email protected]> * fix(googlechat): bound api error body reads --------- Co-authored-by: Claude <[email protected]> Co-authored-by: Vincent Koc <[email protected]>

…sonResponse (openclaw#96772) * fix(googlechat): replace unbounded response.json() with readProviderJsonResponse Replace the local readGoogleChatJsonResponse and readGoogleChatCertsResponse wrappers with the existing SDK helper readProviderJsonResponse (from openclaw/plugin-sdk/provider-http) so the Google Chat API JSON responses are bounded at 16 MiB, matching the non-streaming cap already used by 15+ other extensions. What changed: - extensions/googlechat/src/api.ts: readGoogleChatJsonResponse now delegates to readProviderJsonResponse. Removed the local try/catch wrapper. - extensions/googlechat/src/auth.ts: readGoogleChatCertsResponse now delegates to readProviderJsonResponse. Error message preserved. Removed the local try/catch wrapper. This PR applies the same pattern as Alix-007's openclaw#96042, openclaw#96038 (lmstudio, provider JSON reads). No SDK promotion needed — readProviderJsonResponse is already available in openclaw/plugin-sdk/provider-http. * fix(googlechat): add inline bounded-read regression tests Co-Authored-By: Claude <[email protected]> * fix(googlechat): remove unused variable flagged by oxlint Co-Authored-By: Claude <[email protected]> * fix(googlechat): bound api error body reads --------- Co-authored-by: Claude <[email protected]> Co-authored-by: Vincent Koc <[email protected]>

…penclaw#96042) The /api/v1/models/load success path read the response with an unbounded await response.json(), so a misbehaving or compromised LM Studio server could stream an arbitrarily large JSON body that is fully buffered into memory before any size check. Read it through the shared byte-capped readProviderJsonResponse helper instead (16 MiB provider-JSON cap, cancels the stream on overflow, wraps malformed JSON), matching the discovery path and the already-bounded error body. Migrate the model fetch/load test mocks to real Response objects (the bounded readers need a real body stream) and add a regression test that streams an oversized success body and asserts a bounded error plus stream cancellation. Label: security

…sonResponse (openclaw#96772) * fix(googlechat): replace unbounded response.json() with readProviderJsonResponse Replace the local readGoogleChatJsonResponse and readGoogleChatCertsResponse wrappers with the existing SDK helper readProviderJsonResponse (from openclaw/plugin-sdk/provider-http) so the Google Chat API JSON responses are bounded at 16 MiB, matching the non-streaming cap already used by 15+ other extensions. What changed: - extensions/googlechat/src/api.ts: readGoogleChatJsonResponse now delegates to readProviderJsonResponse. Removed the local try/catch wrapper. - extensions/googlechat/src/auth.ts: readGoogleChatCertsResponse now delegates to readProviderJsonResponse. Error message preserved. Removed the local try/catch wrapper. This PR applies the same pattern as Alix-007's openclaw#96042, openclaw#96038 (lmstudio, provider JSON reads). No SDK promotion needed — readProviderJsonResponse is already available in openclaw/plugin-sdk/provider-http. * fix(googlechat): add inline bounded-read regression tests Co-Authored-By: Claude <[email protected]> * fix(googlechat): remove unused variable flagged by oxlint Co-Authored-By: Claude <[email protected]> * fix(googlechat): bound api error body reads --------- Co-authored-by: Claude <[email protected]> Co-authored-by: Vincent Koc <[email protected]>

The fal music generation provider in extensions/fal/music-generation-provider.ts parsed its HTTP success response with an unbounded await response.json(). A hostile or buggy fal.ai endpoint — the base URL is user-configurable via models.providers.fal.baseUrl (resolved by resolveFalHttpRequestConfig in http-config.ts, defaulting to https://fal.run) — could stream an arbitrarily large JSON body into memory before parsing, forcing the music generation handler into OOM. Route the read through the shared readProviderJsonResponse (from openclaw/plugin-sdk/provider-http), which enforces the 16 MiB byte cap (PROVIDER_JSON_RESPONSE_MAX_BYTES), cancels the stream on overflow, and wraps malformed JSON with the caller label. The change is minimal — a single .json() call replaced by readProviderJsonResponse<unknown> — and preserves the existing downstream logic unchanged. Update the test suite: replace fake { json: async () => ... } response stubs with proper Response objects carrying JSON body streams via a new jsonBodyResponse() helper, so the real readProviderJsonResponse implementation actually exercises the byte-bounded reader under test. Preserve the real readProviderJsonResponse in the provider-http mock (via importOriginal) so the bounded reader streams and cancels oversized bodies. Add a focused regression test: when the music generation stream exceeds the JSON byte cap (32 MiB body, double the 16 MiB cap), generateMusic rejects with a "fal-music-generation: JSON response exceeds" error and the reader cancels the body mid-flight (ReadableStream.cancel fires, bytesPulled < 32 MiB). Existing parse/HTTP-error cases keep passing. Symmetric counterpart to the openclaw#96027/openclaw#96038/openclaw#96042/openclaw#96606/openclaw#96607 response-limit campaign.

…sonResponse (openclaw#96772) * fix(googlechat): replace unbounded response.json() with readProviderJsonResponse Replace the local readGoogleChatJsonResponse and readGoogleChatCertsResponse wrappers with the existing SDK helper readProviderJsonResponse (from openclaw/plugin-sdk/provider-http) so the Google Chat API JSON responses are bounded at 16 MiB, matching the non-streaming cap already used by 15+ other extensions. What changed: - extensions/googlechat/src/api.ts: readGoogleChatJsonResponse now delegates to readProviderJsonResponse. Removed the local try/catch wrapper. - extensions/googlechat/src/auth.ts: readGoogleChatCertsResponse now delegates to readProviderJsonResponse. Error message preserved. Removed the local try/catch wrapper. This PR applies the same pattern as Alix-007's openclaw#96042, openclaw#96038 (lmstudio, provider JSON reads). No SDK promotion needed — readProviderJsonResponse is already available in openclaw/plugin-sdk/provider-http. * fix(googlechat): add inline bounded-read regression tests Co-Authored-By: Claude <[email protected]> * fix(googlechat): remove unused variable flagged by oxlint Co-Authored-By: Claude <[email protected]> * fix(googlechat): bound api error body reads --------- Co-authored-by: Claude <[email protected]> Co-authored-by: Vincent Koc <[email protected]> (cherry picked from commit 4f3d81b)

openclaw-barnacle Bot added extensions: lmstudio size: S labels Jun 23, 2026

sallyom self-assigned this Jun 24, 2026

sallyom merged commit 7844b08 into openclaw:main Jun 24, 2026
132 of 139 checks passed

github-actions Bot mentioned this pull request Jun 24, 2026

📡 Upstream Digest — 2026-06-24 14:10 UTC curtismercier/openclaw-mods#1144

Open

This was referenced Jun 24, 2026

fix(exa): bound untrusted search JSON response reads #96038

Merged

fix(parallel): bound successful web-search JSON response reads #96035

Merged

fix(ollama): bound model-discovery JSON response reads #96027

Merged

This was referenced Jun 25, 2026

fix(anthropic): bound streaming 200 success-body SSE reads via shared plugin-sdk guard #96632

Closed

fix(sdk): promote streaming-byte-guard to plugin-sdk + bound 4 core streaming reads #96688

Closed

clawsweeper Bot mentioned this pull request Jun 25, 2026

fix(googlechat): replace unbounded response.json() with readProviderJsonResponse #96772

Merged

hugenshen mentioned this pull request Jul 7, 2026

fix(mattermost): truncate inbound preview on code-point boundary #101630

Merged

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Uh oh!

Uh oh!

fix(lmstudio): bound model load success response body to prevent OOM#96042

fix(lmstudio): bound model load success response body to prevent OOM#96042
sallyom merged 1 commit into
openclaw:mainfrom
Alix-007:fix/bound-lmstudio-model-fetch

Alix-007 commented Jun 23, 2026

Uh oh!

Alix-007 commented Jun 23, 2026

Uh oh!

clawsweeper Bot commented Jun 23, 2026 •

edited

Loading

Uh oh!

clawsweeper Bot commented Jun 23, 2026 •

edited

Loading

Uh oh!

sallyom commented Jun 24, 2026

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

Uh oh!

Uh oh!

Conversation

Alix-007 commented Jun 23, 2026

What Problem This Solves

Changes

Real behavior proof

Evidence

Uh oh!

Alix-007 commented Jun 23, 2026

Uh oh!

clawsweeper Bot commented Jun 23, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

clawsweeper Bot commented Jun 23, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

sallyom commented Jun 24, 2026

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

clawsweeper Bot commented Jun 23, 2026 •

edited

Loading

clawsweeper Bot commented Jun 23, 2026 •

edited

Loading