fix(qa-lab): use truncateUtf16Safe for tool search gateway and mock OpenAI text truncation by lsr911 · Pull Request #103219 · openclaw/openclaw

lsr911 · 2026-07-10T01:19:42Z

What Problem This Solves

QA Lab's tool-search diagnostics and mock OpenAI response builder truncated text with raw UTF-16 slicing at 220, 300, 500, and 4,000-code-unit boundaries. Emoji and other non-BMP characters at those boundaries could leave lone surrogate halves in debug evidence or mock responses.

Why This Change Was Made

The five bounded prefix sites now use the canonical truncateUtf16Safe helper exposed through the public plugin SDK, following the cross-repository pattern established by #102823. The standalone helper proof was replaced with regression tests through the real tool-search lane, assertion diagnostics, and mock HTTP response paths.

User Impact

QA Lab failure diagnostics and mock OpenAI evidence responses remain well-formed when long provider, tool, or gateway text contains emoji or other non-BMP characters. Existing code-unit limits and scenario behavior are unchanged.

Evidence

Exact-head isolated proof at b6c6b426a6f5600582b646a438668ad56e44a46c: AWS Crabbox run run_ebcb4ec91b63 passed both focused files (136 tests total).
The same exact-head run built the runtime and passed the full tool-search-gateway-e2e mock-provider scenario (1/1).
Targeted oxfmt --check passed for all four changed files.
Fresh exact-head autoreview: clean, confidence 0.98.

clawsweeper · 2026-07-10T09:25:32Z

Codex review: needs real behavior proof before merge. Reviewed July 10, 2026, 6:22 PM ET / 22:22 UTC.

Summary
The PR replaces five QA Lab raw prefix slices with truncateUtf16Safe and adds integrated surrogate-boundary regressions through the tool-search fixture and mock OpenAI HTTP server.

PR surface: Source +4, Tests +124. Total +128 across 4 files.

Reproducibility: yes. from source: placing a supplementary character one UTF-16 code unit before any current cap causes raw .slice(0, N) to return a lone high surrogate. This review did not execute the tests.

Review metrics: none identified.

Stored data model
Persistent data-model change detected: serialized state: extensions/qa-lab/src/tool-search-gateway.fixture.test.ts. Confirm migration or upgrade compatibility proof before merge.

Root-cause cluster
Relationship: canonical
Canonical: #103219
Summary: This PR is the canonical remaining fix for these exact QA Lab boundaries; merged related work addresses the same defect family on other surfaces.

Members:

partial_overlap: fix: keep bounded Unicode text valid across UTF-16 boundaries #102823 - The merged repository-wide sweep established the canonical safe-truncation pattern, but current main still contains the five exact unsafe boundaries changed here.
adjacent_distinct: fix(qa-matrix): keep shared reply previews UTF-16 safe #102656 - The merged PR fixed the same UTF-16 boundary class in a separate Matrix QA reply-artifact path.

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

Merge readiness
Overall: 🦪 silver shellfish
Proof: 🧂 unranked krab
Patch quality: 🐚 platinum hermit
Result: blocked until real behavior proof from a real setup is added.

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

Rank-up moves:

Attach redacted exact-head QA Lab runtime output or logs demonstrating the changed snippet paths with an emoji crossing each relevant boundary.

Proof guidance:

[P1] Needs real behavior proof before merge: The current head has useful integrated mock and HTTP regressions, but no after-fix exact-head real QA Lab run, logs, or live output; add redacted evidence, update the PR body to trigger a fresh review, or ask a maintainer to comment @clawsweeper re-review.

Risk before merge

[P1] The PR discussion does not contain an after-fix exact-head execution of the changed QA Lab workflow; integrated tests and mocked HTTP paths are supplemental and do not satisfy the external-contributor real-behavior proof requirement.

Maintainer options:

Decide the mitigation before merge
Keep the canonical helper at all five truncation boundaries with the integrated regressions, then attach redacted exact-head QA Lab runtime output or logs before merge.
Pause or close
Do not merge this PR until maintainers decide whether the risk is worth taking.

Next step before merge

[P1] No actionable code defect remains established; contributor-supplied exact-head real-behavior proof is the only identified merge blocker.

Security
Cleared: The QA Lab runtime and test changes add no dependency, workflow, permission, secret, install, publishing, or package-resolution surface.

Review details

Best possible solution:

Keep the canonical helper at all five truncation boundaries with the integrated regressions, then attach redacted exact-head QA Lab runtime output or logs before merge.

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

Yes from source: placing a supplementary character one UTF-16 code unit before any current cap causes raw .slice(0, N) to return a lone high surrogate. This review did not execute the tests.

Is this the best way to solve the issue?

Yes; applying the existing shared helper at each owning truncation boundary preserves the established UTF-16-unit budgets and avoids introducing a parallel Unicode policy.

AGENTS.md: found and applied where relevant.

Codex review notes: model internal, reasoning high; reviewed against 53bcdf8fe25f.

Label changes

Label justifications:

P3: The fix protects internal QA evidence and diagnostics from malformed Unicode without changing production config, protocol, persisted state, or defaults.
rating: 🦪 silver shellfish: Overall readiness is 🦪 silver shellfish; proof is 🧂 unranked krab and patch quality is 🐚 platinum hermit.
status: 📣 needs proof: The PR needs real behavior proof before ClawSweeper can clear the contributor ask. Needs real behavior proof before merge: The current head has useful integrated mock and HTTP regressions, but no after-fix exact-head real QA Lab run, logs, or live output; add redacted evidence, update the PR body to trigger a fresh review, or ask a maintainer to comment @clawsweeper re-review.

Evidence reviewed

PR surface:

Source +4, Tests +124. Total +128 across 4 files.

View PR surface stats

Area	Files	Added	Removed	Net
Source	2	10	6	+4
Tests	2	125	1	+124
Docs	0	0	0	0
Config	0	0	0	0
Generated	0	0	0	0
Other	0	0	0	0
Total	4	135	7	+128

What I checked:

Current main remains affected: Current main still applies raw .slice(0, N) to provider input, provider tool output, both gateway debug outputs, and the mock OpenAI evidence snippet. (extensions/qa-lab/src/tool-search-gateway.fixture.ts:427, 53bcdf8fe25f)
Implementation uses the canonical helper: The branch applies the public plugin SDK helper at the existing truncation boundaries while preserving the 500, 4000, and 300 UTF-16-unit limits. (extensions/qa-lab/src/tool-search-gateway.fixture.ts:428, b6c6b426a6f5)
Prior review finding resolved: The latest head adds regressions through runToolSearchGatewayLane, assertToolSearchLaneResults, and the mock OpenAI HTTP server, replacing the earlier helper-only proof approach. (extensions/qa-lab/src/tool-search-gateway.fixture.test.ts:149, b6c6b426a6f5)
Mock server path has integrated coverage: The regression starts the mock HTTP server, submits a function-call output whose emoji crosses the 220-unit boundary, and asserts the returned evidence text has no split surrogate. (extensions/qa-lab/src/providers/mock-openai/server.test.ts:834, b6c6b426a6f5)
Feature history provenance: Blame ties the affected QA Lab fixture and mock-server paths to the original QA Lab implementation commit; the current head's integrated coverage was subsequently authored in commit b6c6b42. (extensions/qa-lab/src/providers/mock-openai/server.ts:1713, 8a93d288e2f5)
Exact-head CI: The current head reports 69 successful checks, with two checks still pending at review time and no observed failure. (b6c6b426a6f5)

Likely related people:

steipete: Authored the current-head integrated boundary coverage and the recent merged repository-wide UTF-16 truncation sweep. (role: recent area contributor; confidence: high; commits: b6c6b426a6f5, 5ef269c2bbfc; files: extensions/qa-lab/src/tool-search-gateway.fixture.test.ts, extensions/qa-lab/src/providers/mock-openai/server.test.ts, extensions/qa-lab/src/tool-search-gateway.fixture.ts)
pick-cat: Git history attributes the central QA Lab fixture and mock-server truncation paths to the original implementation commit. (role: introduced behavior; confidence: medium; commits: 8a93d288e2f5; files: extensions/qa-lab/src/tool-search-gateway.fixture.ts, extensions/qa-lab/src/providers/mock-openai/server.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.

Review history (2 earlier review cycles)

reviewed 2026-07-10T09:23:56.365Z sha c376504 :: needs real behavior proof before merge. :: none
reviewed 2026-07-10T12:13:16.707Z sha 03d39e6 :: needs real behavior proof before merge. :: [P2] Make the UTF-16 regression proof exercise the changed paths

lsr911 · 2026-07-10T10:08:43Z

@clawsweeper re-review

clawsweeper · 2026-07-10T10:08:46Z

🦞🧹
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/29091555552
Updated: 2026-07-10T12:15:01.621Z

…xt truncation Replace naive .slice(0, N) with truncateUtf16Safe() in: - tool-search-gateway.fixture.ts: 4 sites (provider input snippet, tool output snippet, gateway output text for normal/code lanes) - mock-openai/server.ts: 1 site (tool output evidence snippet) LLM output text may contain non-BMP characters (emoji, CJK extension characters). Naive .slice(0, N) at a surrogate pair boundary produces a lone surrogate. Co-Authored-By: Claude <[email protected]>

…aries

Co-authored-by: lsr911 <[email protected]>

steipete · 2026-07-10T22:26:02Z

Land-ready review complete at exact head b6c6b426a6f5600582b646a438668ad56e44a46c.

Best-fix review: all five scoped prefix boundaries now use the public canonical helper. The disposable helper-only proof was removed in favor of tests through the real tool-search lane, assertion diagnostics, and mock HTTP response path. No findings or nits remain.
Hosted exact-head CI: all relevant checks passed with no failures or pending jobs.
Isolated fork-safe proof: AWS Crabbox cbx_0c012f520af7, run run_ebcb4ec91b63, public network with no Tailscale or hydrated credentials. Both focused files passed 136/136 tests; the full tool-search-gateway-e2e mock-provider scenario passed 1/1 after a runtime build.
Targeted formatting passed for all four changed files.
Fresh exact-head autoreview: clean, confidence 0.98.
Native prepare-run: passed with hosted exact-head gates; prepared SHA equals the PR head.

No known proof gaps.

steipete · 2026-07-10T22:26:46Z

Merged via squash.

Prepared head SHA: b6c6b426a6f5600582b646a438668ad56e44a46c
Landed commit: 9dee9ebffa582d6191c59f307ffdd1bfddfc1a2f

…penAI text truncation (openclaw#103219) * fix(qa-lab): use truncateUtf16Safe for tool search and mock OpenAI text truncation Replace naive .slice(0, N) with truncateUtf16Safe() in: - tool-search-gateway.fixture.ts: 4 sites (provider input snippet, tool output snippet, gateway output text for normal/code lanes) - mock-openai/server.ts: 1 site (tool output evidence snippet) LLM output text may contain non-BMP characters (emoji, CJK extension characters). Naive .slice(0, N) at a surrogate pair boundary produces a lone surrogate. Co-Authored-By: Claude <[email protected]> * test: add proof script for qa-lab UTF-16 safe truncation at all boundaries * test(qa-lab): integrate UTF-16 boundary coverage Co-authored-by: lsr911 <[email protected]> --------- Co-authored-by: Claude <[email protected]> Co-authored-by: Peter Steinberger <[email protected]>

openclaw-barnacle Bot added extensions: qa-lab size: XS labels Jul 10, 2026

openclaw-barnacle Bot added size: S and removed size: XS labels Jul 10, 2026

clawsweeper Bot added rating: 🦪 silver shellfish Thin PR readiness signal; proof, validation, or implementation needs work. and removed rating: 🧂 unranked krab Not merge-ready due to missing proof or serious correctness/safety concerns. labels Jul 10, 2026

lsr911 and others added 3 commits July 10, 2026 23:14

test: add proof script for qa-lab UTF-16 safe truncation at all bound…

804a78f

…aries

test(qa-lab): integrate UTF-16 boundary coverage

b6c6b42

Co-authored-by: lsr911 <[email protected]>

steipete force-pushed the fix/qa-lab-utf16-truncation branch from 03d39e6 to b6c6b42 Compare July 10, 2026 22:15

steipete merged commit 9dee9eb into openclaw:main Jul 10, 2026
103 checks passed

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

📡 Upstream Digest — 2026-07-10 22:52 UTC curtismercier/openclaw-mods#1255

Open

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Uh oh!

Uh oh!

fix(qa-lab): use truncateUtf16Safe for tool search gateway and mock OpenAI text truncation#103219

fix(qa-lab): use truncateUtf16Safe for tool search gateway and mock OpenAI text truncation#103219
steipete merged 3 commits into
openclaw:mainfrom
lsr911:fix/qa-lab-utf16-truncation

lsr911 commented Jul 10, 2026 •

edited by steipete

Loading

Uh oh!

clawsweeper Bot commented Jul 10, 2026 •

edited

Loading

Uh oh!

lsr911 commented Jul 10, 2026

Uh oh!

clawsweeper Bot commented Jul 10, 2026 •

edited

Loading

Uh oh!

steipete commented Jul 10, 2026

Uh oh!

Uh oh!

steipete commented Jul 10, 2026

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

Uh oh!

Uh oh!

Conversation

lsr911 commented Jul 10, 2026 • edited by steipete 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 Jul 10, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

lsr911 commented Jul 10, 2026

Uh oh!

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

Uh oh!

Uh oh!

steipete commented Jul 10, 2026

Uh oh!

Uh oh!

steipete commented Jul 10, 2026

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

lsr911 commented Jul 10, 2026 •

edited by steipete

Loading

clawsweeper Bot commented Jul 10, 2026 •

edited

Loading

clawsweeper Bot commented Jul 10, 2026 •

edited

Loading