Codex review: needs maintainer review before merge. Reviewed May 27, 2026, 1:00 AM ET / 05:00 UTC.

Summary
The PR eagerly writes WebChat chat.send user turns to the active session transcript before agent dispatch, propagates Pi suppression metadata, updates WebChat docs, and adds gateway/Pi regression coverage.

PR surface: Source +519, Tests +916, Docs +1. Total +1436 across 17 files.

Reproducibility: yes. at source/test level: the PR body gives a blocked-lane WebChat path, and the added test holds dispatch while checking JSONL append before release. I did not rerun it because this review is read-only.

Review metrics: 1 noteworthy metric.

• Suppression contract fields: 3 added. The PR adds boolean, session-id, and entry-id suppression metadata across reply options and Pi run params, so the internal run contract needs review before merge.

Merge readiness
Overall: 🦐 gold shrimp
Proof: 🐚 platinum hermit
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 onto current main and preserve the current chat.send user-turn recorder and media-offload routing behavior.
• Rerun the focused gateway/Pi regression tests and changed checks after the rebase.

Mantis proof suggestion
A short real WebChat proof would materially show the user-visible latency fix and duplicate-turn behavior. A maintainer can ask Mantis to capture proof by posting a new PR comment that starts with the OpenClaw Mantis account mention, followed by:

visual task: verify WebChat shows a sent user turn immediately after chat.send ACK while an agent lane is blocked, with no duplicate user turn after reply.

Risk before merge

• The PR is reported dirty/unmergeable against current main, and the conflict area includes newer chat.send user-turn recorder and media-offload handling that must not be lost during rebase.
• The patch moves durable user-turn persistence ahead of agent dispatch and coordinates suppression through Pi, so mistakes can duplicate, omit, or mis-associate WebChat transcript entries across rollover, hooks, and queued same-session runs.
• The supplied proof is credible for the reporter's local gateway, but the PR body notes that a freshly built current-main Control UI was not tested against that running gateway because of protocol mismatch.

Maintainer options:

1. Rebase and preserve current-main transcript behavior (recommended)
   Resolve the dirty branch by replaying the eager append design on top of current main without dropping the user-turn recorder or text-only media offload routing.
2. Accept owner-reviewed persistence risk
   A maintainer can choose to land the eager transcript behavior after explicitly owning the session rollover, hook, and duplicate-suppression risk.

Next step before merge
Manual review is appropriate because the branch is dirty and the remaining work is integration and session-state safety review, not a narrow automated repair.

Security
Cleared: No dependency, workflow, permission, or secret-handling surface was added; the reviewed branch keeps redaction and hook gates on the new transcript path.

Maintainer merge request for head $sha.

Summary:

• Kept the WebChat eager user transcript fix, so chat.send persists the user turn before entering the agent/session lane.
• Fixed the attachment regression from the eager path: attachment media persistence no longer blocks dispatch, and the live session.message is emitted with media metadata only after the media rewrite completes.

After-fix verification from local OpenClaw source checkout:

ode scripts/test-projects.mjs src/gateway/server-methods/chat.directive-tags.test.ts -> 73 passed.

ode scripts/test-projects.mjs src/agents/pi-embedded-runner/run/attempt.prompt-helpers.test.ts src/gateway/server.chat.gateway-server-chat.test.ts src/gateway/server.chat.gateway-server-chat-b.test.ts src/auto-reply/reply/agent-runner-utils.test.ts src/auto-reply/reply/agent-runner-execution.test.ts src/auto-reply/reply/followup-runner.test.ts src/agents/pi-embedded-runner/run.overflow-compaction.loop.test.ts -> 7 files / 195 tests passed.

ode_modules.bin\oxfmt.CMD --check --threads=1 src\gateway\server-methods\chat.ts -> passed.

• corepack pnpm tsgo:core -> passed.
• corepack pnpm tsgo:core:test -> passed.
• corepack pnpm check:changed -> passed.
• git diff --check and git diff --check origin/main..HEAD -> passed.

GitHub checks observed on the latest SHA:

• Real behavior proof -> success.
• label, label-issues, and dispatch -> success.
• One �uto-response check was cancelled for the superseded push; the rerun on the latest SHA completed successfully.

Please merge when maintainer-required CI/review is satisfied.

Update after rebasing onto latest origin/main (839e9e9): head is now $sha.

Mergeability:

• Rebased cleanly after resolving the only conflict in CHANGELOG.md by dropping the contributor changelog entry, per repository guidance that maintainer/AI handles changelog during landing.
• GitHub now reports the PR as mergeable (mergeable=true); current unstable state is pending CI on the new SHA.

Re-run local verification after rebase:

ode scripts/test-projects.mjs src/gateway/server-methods/chat.directive-tags.test.ts -> 73 passed.

ode scripts/test-projects.mjs src/agents/pi-embedded-runner/run/attempt.prompt-helpers.test.ts src/gateway/server.chat.gateway-server-chat.test.ts src/gateway/server.chat.gateway-server-chat-b.test.ts src/auto-reply/reply/agent-runner-utils.test.ts src/auto-reply/reply/agent-runner-execution.test.ts src/auto-reply/reply/followup-runner.test.ts src/agents/pi-embedded-runner/run.overflow-compaction.loop.test.ts -> 7 files / 195 tests passed.

ode_modules.bin\oxfmt.CMD --check --threads=1 src\gateway\server-methods\chat.ts src\gateway\server-methods\chat-user-transcript-persistence.ts src\gateway\server.chat.gateway-server-chat-b.test.ts src\agents\pi-embedded-runner\run\attempt.prompt-helpers.ts src\agents\pi-embedded-runner\run\attempt.prompt-helpers.test.ts -> passed.

• corepack pnpm tsgo:core -> passed.
• corepack pnpm check:changed -> passed.
• git diff --check origin/main..HEAD -> passed.

Known local check gap:

• corepack pnpm tsgo:core:test currently fails on latest main in unrelated ACP tests: src/acp/translator.session-rate-limit.test.ts and src/acp/translator.stop-reason.test.ts. This PR has no src/acp/** changes, so I did not expand this latency fix into ACP test typing cleanup.

Please merge once required CI/review is satisfied.

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/25686750806
• Updated: 2026-05-11T17:45:05.037Z

CI follow-up for the latest head (c3f576d54342c54cfffcddcdb3d1051cc8860971):

• Fixed the checks-node-agentic-gateway-methods failure in chat.directive-tags.test.ts by waiting for the intentionally backgrounded eager media transcript path to drain before the test exits. The previous assertion only waited for media saves, so a late transcript update from "serial please" could leak into the following "quick command" test on CI.
• This is test-only; the gateway behavior remains the same.

Local verification after the fix:

node scripts/test-projects.mjs src/gateway/server-methods/chat.directive-tags.test.ts
# 1 file / 73 tests passed

node scripts/test-projects.mjs src/gateway/server-methods/chat.directive-tags.test.ts src/gateway/server.chat.gateway-server-chat.test.ts src/gateway/server.chat.gateway-server-chat-b.test.ts
# 3 files / 122 tests passed

node_modules\.bin\oxfmt.CMD --check --threads=1 src/gateway/server-methods/chat.directive-tags.test.ts
# passed

corepack pnpm check:changed
# passed

git diff --check origin/main..HEAD
git diff --check
# passed

I also attempted the full CI shard locally with test/vitest/vitest.gateway-methods.config.ts, but the Windows local run exceeded a 5 minute timeout before producing a result, so I am relying on the targeted gateway-method tests above plus the GitHub shard rerun.

Maintainer merge request update:

• The previous Real behavior proof blocker is fixed and passing on the latest head.
• The follow-up checks-node-agentic-gateway-methods failure is fixed by c3f576d54342c54cfffcddcdb3d1051cc8860971; the GitHub shard is now passing.
• Current state has no failed checks; remaining checks are still queued in GitHub Actions.

Please merge once the required queued checks complete, unless another maintainer concern comes up.

This pull request has been automatically marked as stale due to inactivity.
Please add updates or it will be closed.

ClawSweeper PR egg

✨ Hatched: 🥚 common Pearl Diff Drake

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: 🥚 common.
Trait: polishes edge cases.
Image traits: location green-check meadow; accessory tiny test log scroll; palette violet, aqua, and starlight; mood watchful; pose leaning over a miniature review desk; shell starlit enamel shell; lighting moonlit rim light; background small green status lights.
Share on X: post this hatch
Copy: My PR egg hatched a 🥚 common Pearl Diff Drake in ClawSweeper.