fix(reasoning-tags): strip MiniMax mm: tags on silent-reply and streaming paths missed by #93767 by Alix-007 · Pull Request #93806 · openclaw/openclaw

Alix-007 · 2026-06-16T23:38:30Z

Summary

#93767 added MiniMax mm: namespace support to reasoning-tag handling across 9 files (so <mm:think> etc. are treated like the existing antml:/bare reasoning tags). Two production-path reasoning-tag regexes were missed and still only match antml:/bare tags:

src/auto-reply/tokens.ts:88-91 — taggedReasoningPrefixRe / openReasoningPrefixRe, which feed stripLeadingReasoningBlocks and isReasoningPrefixedSilentReplyText → the exported isSilentReplyPayloadText (used at 14+ production call sites: normalize-reply, reply-directives, pending-final-delivery, replay-history, subagent-announce-delivery, qqbot, …).
src/agents/embedded-agent-subscribe.handlers.messages.ts:216 — REASONING_TAG_RE, the gate at line 737 that decides whether a streaming chunk is recomputed to strip thinking.

Result: a MiniMax model's <mm:think>…</mm:think> reasoning leaks past silent-reply detection and into the visible stream, the exact thing #93767 set out to prevent.

Changes

Both regexes: (?:antml:)? → (?:antml:|mm:)?, character-for-character matching fix(reasoning-tags): strip MiniMax mm: namespaced reasoning tags #93767's pattern (commit 4f860bfab0). +3/-3 across 2 files; pure recognition addition, no other regex/logic change.

Real behavior proof

Behavior addressed: MiniMax mm:-namespaced reasoning tags are now stripped/detected on both the silent-reply path and the streaming-recompute path, matching antml: behavior.

Real environment: local, Node v22, node --import tsx, driving the real exported isSilentReplyPayloadText and handleMessageUpdate (+ the real mm-aware stream sanitizer) — no mocks.

Evidence (real functions):

[silent]  isSilentReplyPayloadText("<mm:think>silent</mm:think>NO_REPLY")  -> silent (after)
          <think>/bare reasoning unchanged; "<mm:think>x</mm:think>Here is the answer: 42." NOT misclassified silent
[stream]  handleMessageUpdate(text_delta "<mm:think>secret</mm:think>Visible answer.") -> emits "Visible answer." (no "secret")

Negative controls (pre-#93767 antml-only regex, real): <mm:think> block survives .replace (leaks); stream gate .test("<mm:think>…") === false (not stripped → leaks).

What was not tested: no live MiniMax provider call; the change is in the deterministic reasoning-tag recognition path, driven via the real exported functions.

Tests and validation

tokens.test.ts + reasoning-tags.test.ts + reasoning-tag-text-partitioner.test.ts + embedded-agent-subscribe.handlers.messages.test.ts: 199 pass.
Negative control reproduces the leak without the fix.
oxlint + oxfmt --check clean on both files.

clawsweeper · 2026-06-16T23:58:06Z

ClawSweeper review: did not complete due to Codex infrastructure failure. Reviewed June 16, 2026, 8:18 PM ET / 00:18 UTC.

Summary
Review failed before ClawSweeper could summarize the requested change.

PR surface: Source 0. Total 0 across 2 files.

Reproducibility: unclear. The review failed before ClawSweeper could establish a reproduction path.

Review metrics: none identified.

Merge readiness
Not assessed.
Failure reason: timeout.

This is a ClawSweeper/Codex infrastructure failure, not a PR readiness or patch-quality verdict.
Keep any merge decision on the normal maintainer review path until ClawSweeper can complete a fresh review.

Risk before merge

[P1] No close action taken because the review did not complete.

Maintainer options:

Decide the mitigation before merge
Retry the Codex review after fixing the execution failure.
Pause or close
Do not merge this PR until maintainers decide whether the risk is worth taking.

Next step before merge

Review did not complete, so no work-lane recommendation was made.

Review details

Best possible solution:

Retry the Codex review after fixing the execution failure.

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

Unclear. The review failed before ClawSweeper could establish a reproduction path.

Is this the best way to solve the issue?

Unclear. Retry the review first so ClawSweeper can evaluate the actual issue and fix direction.

AGENTS.md: unclear because the file could not be read completely.

Codex review notes: model internal, reasoning high; reviewed against 7c97c6da9bfa.

Evidence reviewed

PR surface:

Source 0. Total 0 across 2 files.

View PR surface stats

Area	Files	Added	Removed
Source	2	3	3
Tests	0	0	0
Docs	0	0	0
Config	0	0	0
Generated	0	0	0
Other	0	0	0
Total	2	3	3

What I checked:

failure reason: timeout.
codex failure detail: Codex review failed for this PR: Codex process timed out after 1199999ms.
codex stderr: No stderr captured.
codex stdout: {"type":"thread.started","thread_id":"019ed2de-f902-7882-846e-3678be49dde1"}.
process error code: ETIMEDOUT.

Likely related people:

unknown: Codex failed before it could trace repository history. (role: review did not complete; confidence: low)

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.

…d stream gates PR openclaw#93767 added MiniMax `mm:`-namespaced reasoning-tag support across the shared sanitizer and Telegram lane coordinator, but two production reasoning-tag recognizers were missed and still only matched the `antml:` namespace: - src/auto-reply/tokens.ts: `taggedReasoningPrefixRe` / `openReasoningPrefixRe` drive `stripLeadingReasoningBlocks` and `isSilentReplyPayloadText`, which 14+ call sites use to detect NO_REPLY silent payloads. A `<mm:think>…</mm:think>NO_REPLY` reply was not recognized as silent, leaking the wrapper into delivery. - src/agents/embedded-agent-subscribe.handlers.messages.ts: `REASONING_TAG_RE` gates `shouldRecomputeFullStream`. A `<mm:think>` streaming chunk failed the test, so the visible stream was not recomputed and the hidden reasoning leaked. Add the `mm:` alternative alongside `antml:` in all three regexes, matching the exact `(?:antml:|mm:)?` form used by openclaw#93767. Identification-only change, no other regex logic touched.

vincentkoc · 2026-06-17T01:37:09Z

Maintainer repair and verification complete.

Added focused regressions for MiniMax mm:think handling on both silent-reply detection and the streaming replacement path, including split orphan closing tags and unterminated reasoning wrappers before NO_REPLY.

Verification:

node scripts/run-vitest.mjs src/auto-reply/tokens.test.ts src/agents/embedded-agent-subscribe.subscribe-embedded-agent-session.subscribeembeddedagentsession.test.ts — 96 focused tests passed
final Codex autoreview on the current rebased head — clean
prior changed-gate run passed the real changed checks and stopped only on the delegated checkout's warning-only no-merge-base temp report

Alix-007 · 2026-06-17T02:02:56Z

Thanks for the quick reviews and merges across this recent batch of fixes, @vincentkoc — much appreciated! 🙏

…ming paths missed by openclaw#93767 (openclaw#93806) * fix(reasoning-tags): accept MiniMax mm: prefix in silent-detection and stream gates PR openclaw#93767 added MiniMax `mm:`-namespaced reasoning-tag support across the shared sanitizer and Telegram lane coordinator, but two production reasoning-tag recognizers were missed and still only matched the `antml:` namespace: - src/auto-reply/tokens.ts: `taggedReasoningPrefixRe` / `openReasoningPrefixRe` drive `stripLeadingReasoningBlocks` and `isSilentReplyPayloadText`, which 14+ call sites use to detect NO_REPLY silent payloads. A `<mm:think>…</mm:think>NO_REPLY` reply was not recognized as silent, leaking the wrapper into delivery. - src/agents/embedded-agent-subscribe.handlers.messages.ts: `REASONING_TAG_RE` gates `shouldRecomputeFullStream`. A `<mm:think>` streaming chunk failed the test, so the visible stream was not recomputed and the hidden reasoning leaked. Add the `mm:` alternative alongside `antml:` in all three regexes, matching the exact `(?:antml:|mm:)?` form used by openclaw#93767. Identification-only change, no other regex logic touched. * test(agents): cover MiniMax reasoning regressions --------- Co-authored-by: Vincent Koc <[email protected]>

…eplies The THINKING_TAG_RE and THINKING_BLOCK_RE regex patterns in the Matrix monitor replies module were missing the mm: namespace prefix for detecting MiniMax reasoning tags. This caused mm:think/mm:thought reasoning content to leak into Matrix room messages. Add (?:antml:|mm:)? prefix to think(?:ing)?|thought alternatives in both regexes, consistent with the pattern already applied in reasoning-tags.ts and other channel modules (see openclaw#93767, openclaw#93806, openclaw#93820). Fixes openclaw#99412

openclaw-barnacle Bot added agents Agent runtime and tooling size: XS triage: needs-real-behavior-proof Candidate: external PR needs after-fix proof from a real setup. labels Jun 16, 2026

vincentkoc self-assigned this Jun 17, 2026

Alix-007 force-pushed the fix/minimax-mm-reasoning-tag-siblings branch from c27b26d to b7de462 Compare June 17, 2026 01:26

Alix-007 force-pushed the fix/minimax-mm-reasoning-tag-siblings branch from b7de462 to 94dd15d Compare June 17, 2026 01:33

test(agents): cover MiniMax reasoning regressions

d4bf2ba

vincentkoc merged commit f8e7d66 into openclaw:main Jun 17, 2026
162 of 165 checks passed

vincentkoc mentioned this pull request Jun 17, 2026

fix(reasoning-tags): strip MiniMax mm: namespaced reasoning tags #93785

Closed

This was referenced Jun 17, 2026

Ecosystem Digest — 2026-06-17 Haderach-Ram/openclaw-radar#49

Open

Ecosystem Digest — 2026-06-17 Haderach-Ram/openclaw-radar#50

Closed

Ecosystem Digest — 2026-06-17 Haderach-Ram/openclaw-radar#51

Closed

This was referenced Jun 17, 2026

fix(slack): recognize MiniMax mm: namespaced reasoning tags in monitor preview #93874

Merged

fix(matrix): recognize MiniMax mm: namespaced reasoning tags in monitor suppression #93926

Closed

zhangguiping-xydt mentioned this pull request Jun 17, 2026

fix(matrix): recognize MiniMax mm: namespaced reasoning tags in monitor replies #94038

Closed

Alix-007 mentioned this pull request Jun 17, 2026

fix(control-ui): recognize namespaced reasoning tags in thinking extraction #94165

Closed

This was referenced Jun 19, 2026

[Bug]: Reasoning tokens leak to chat channels when models stream interleaved text/thinking blocks #89473

Open

fix(imessage): recognize MiniMax mm: reasoning tags in reflection guard (completes #93767) #93820

Merged

sheyanmin mentioned this pull request Jul 3, 2026

fix(matrix): THINKING_TAG_RE missing mm: namespace prefix in monitor replies #99412

Open

krissding mentioned this pull request Jul 3, 2026

fix(matrix): add mm:/antml: namespace prefix to THINKING regex in monitor replies #99415

Closed

sheyanmin mentioned this pull request Jul 3, 2026

fix(matrix): add mm: namespace prefix to THINKING_TAG_RE in monitor replies #99417

Closed

9 tasks

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Uh oh!

Uh oh!

fix(reasoning-tags): strip MiniMax mm: tags on silent-reply and streaming paths missed by #93767#93806

fix(reasoning-tags): strip MiniMax mm: tags on silent-reply and streaming paths missed by #93767#93806
vincentkoc merged 2 commits into
openclaw:mainfrom
Alix-007:fix/minimax-mm-reasoning-tag-siblings

Alix-007 commented Jun 16, 2026

Uh oh!

clawsweeper Bot commented Jun 16, 2026 •

edited

Loading

Uh oh!

vincentkoc commented Jun 17, 2026

Uh oh!

Uh oh!

Alix-007 commented Jun 17, 2026

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

Uh oh!

Uh oh!

Conversation

Alix-007 commented Jun 16, 2026

Summary

Changes

Real behavior proof

Tests and validation

Uh oh!

clawsweeper Bot commented Jun 16, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

vincentkoc commented Jun 17, 2026

Uh oh!

Uh oh!

Alix-007 commented Jun 17, 2026

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

clawsweeper Bot commented Jun 16, 2026 •

edited

Loading