Best possible solution:

Land the existing narrow routing and silent-completion filtering after adding a single-line changelog entry and completing maintainer review.

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

Yes. Source inspection gives a high-confidence path: current main preserves agent thread metadata outside the message tool, but Slack auto-threading only sees toolContext.currentThreadTs plus an enabled reply mode.

Is this the best way to solve the issue?

Mostly yes. Forwarding the existing thread context through the message-tool seam and preserving explicit reply-mode opt-outs is the narrow maintainable fix; the missing changelog entry is the merge-readiness gap.

Full review comments:

• [P3] Add the required changelog entry — src/agents/tools/message-tool.ts:708-710
  This changes user-visible Slack resumed-send threading and subagent completion output, but the PR does not update CHANGELOG.md. OpenClaw policy requires a single-line changelog entry for user-facing fixes before merge.
  Confidence: 0.88

Overall correctness: patch is correct
Overall confidence: 0.84

Acceptance criteria:

• git diff --check
• pnpm test src/agents/subagent-announce-output.test.ts src/agents/tools/message-tool.test.ts --testTimeout=120000

What I checked:

• Current main drops ambient agent thread context: createOpenClawTools forwards current Slack channel/thread fields to createMessageTool, but current main does not pass agentThreadId, so resumed session thread metadata cannot reach the message tool fallback path. (src/agents/openclaw-tools.ts:474, 978bc53e80cc)
• Current main message tool only uses explicit currentThreadTs: The tool context is built from options.currentThreadTs and options.replyToMode; there is no fallback from the agent session thread id on current main. (src/agents/tools/message-tool.ts:834, 978bc53e80cc)
• Slack auto-threading requires thread context and enabled reply mode: Slack's auto-thread resolver returns a thread only when current channel/thread context exists and reply mode permits all or single-use threading. (extensions/slack/src/action-threading.ts:15, 978bc53e80cc)
• Current main renders every child completion row: buildChildCompletionFindings currently pushes every frozen child result into visible findings, so successful ANNOUNCE_SKIP or silent replies can become no-output child completion content. (src/agents/subagent-announce-output.ts:448, 978bc53e80cc)
• PR diff adds the intended wiring and coverage: Provided PR context for head 9d5991f6826b5beacabf828e248b998ff5d6f18f shows agentThreadId forwarded into createMessageTool, computed fallback currentThreadTs/replyToMode, and new message-tool/subagent tests. (src/agents/tools/message-tool.ts:708, 9d5991f6826b)
• Required changelog entry is still absent: The active Unreleased changelog has related Slack/subagent entries, but no entry for this resumed-send threading and silent child completion fix. (CHANGELOG.md:5, 978bc53e80cc)

Likely related people:

• Josh Lehman: Local blame points the current message-tool, subagent completion, and Slack auto-threading code inspected here to commit 6dae3c273de6, which is the current wrapper behavior this PR adjusts. (role: introduced current behavior; confidence: medium; commits: 6dae3c273de6; files: src/agents/tools/message-tool.ts, src/agents/subagent-announce-output.ts, extensions/slack/src/action-threading.ts)
• steipete: Provided ClawSweeper context identifies recent merged work around Slack thread-anchor behavior and subagent completion delivery, which are the two behavioral surfaces touched by this PR. (role: recent maintainer and adjacent owner; confidence: medium; commits: e40d7abda9ff, 42b7b2b9242a, bbef1c555791; files: extensions/slack/src/threading-tool-context.ts, extensions/slack/src/action-threading.ts, src/agents/subagent-announce-output.ts)

Remaining risk / open question:

• The PR author's validation commands were not rerun under this read-only review.
• A live Slack resumed-send scenario was not exercised; the reproduction and fix assessment are source-based.

Codex review notes: model gpt-5.5, reasoning high; reviewed against 978bc53e80cc.