fix: preserve off-mode semantics in auto reply threading (#14976) (thanks @Diaspar4u)

CHANGELOG.md

@@ -7,6 +7,7 @@ Docs: https://docs.openclaw.ai
 ### Fixes
 
 - Security/Audit: distinguish external webhooks (`hooks.enabled`) from internal hooks (`hooks.internal.enabled`) in attack-surface summaries to avoid false exposure signals when only internal hooks are enabled. (#13474) Thanks @mcaxtr.
+- Auto-reply/Threading: auto-inject implicit reply threading so `replyToMode` works without requiring model-emitted `[[reply_to_current]]`, while preserving `replyToMode: "off"` behavior for implicit Slack replies and keeping block-streaming chunk coalescing stable under `replyToMode: "first"`. (#14976) Thanks @Diaspar4u.
 - Sandbox: pass configured `sandbox.docker.env` variables to sandbox containers at `docker create` time. (#15138) Thanks @stevebot-alive.
 - Onboarding/CLI: restore terminal state without resuming paused `stdin`, so onboarding exits cleanly after choosing Web UI and the installer returns instead of appearing stuck.
 - macOS Voice Wake: fix a crash in trigger trimming for CJK/Unicode transcripts by matching and slicing on original-string ranges instead of transformed-string indices. (#11052) Thanks @Flash-LHR.

src/auto-reply/reply/agent-runner-execution.ts

@@ -376,9 +376,11 @@ export async function runAgentTurnWithFallback(params: {
                       text,
                       mediaUrls: payload.mediaUrls,
                       mediaUrl: payload.mediaUrls?.[0],
-                      replyToId: payload.replyToId,
+                      replyToId:
+                        payload.replyToId ??
+                        (payload.replyToCurrent === false ? undefined : currentMessageId),
                       replyToTag: payload.replyToTag,
-                      replyToCurrent: true,
+                      replyToCurrent: payload.replyToCurrent,
                     },
                     currentMessageId,
                   );

src/auto-reply/reply/block-reply-coalescer.ts

@@ -100,10 +100,12 @@ export function createBlockReplyCoalescer(params: {
       return;
     }
 
-    if (
+    const replyToConflict = Boolean(
       bufferText &&
-      (bufferReplyToId !== payload.replyToId || bufferAudioAsVoice !== payload.audioAsVoice)
+      payload.replyToId &&
-    ) {
+      (!bufferReplyToId || bufferReplyToId !== payload.replyToId),
+    );
+    if (bufferText && (replyToConflict || bufferAudioAsVoice !== payload.audioAsVoice)) {
       void flush({ force: true });
     }
 

src/auto-reply/reply/reply-payloads.auto-threading.test.ts

@@ -1,7 +1,7 @@
 import { describe, expect, it } from "vitest";
 import { applyReplyThreading } from "./reply-payloads.js";
 
-describe("applyReplyThreading auto-injects replyToCurrent", () => {
+describe("applyReplyThreading auto-threading", () => {
   it("sets replyToId to currentMessageId even without [[reply_to_current]] tag", () => {
     const result = applyReplyThreading({
       payloads: [{ text: "Hello" }],
@@ -47,4 +47,29 @@ describe("applyReplyThreading auto-injects replyToCurrent", () => {
     expect(result).toHaveLength(1);
     expect(result[0].replyToId).toBeUndefined();
   });
+
+  it("does not bypass off mode for Slack when reply is implicit", () => {
+    const result = applyReplyThreading({
+      payloads: [{ text: "A" }],
+      replyToMode: "off",
+      replyToChannel: "slack",
+      currentMessageId: "42",
+    });
+
+    expect(result).toHaveLength(1);
+    expect(result[0].replyToId).toBeUndefined();
+  });
+
+  it("keeps explicit tags for Slack when off mode allows tags", () => {
+    const result = applyReplyThreading({
+      payloads: [{ text: "[[reply_to_current]]A" }],
+      replyToMode: "off",
+      replyToChannel: "slack",
+      currentMessageId: "42",
+    });
+
+    expect(result).toHaveLength(1);
+    expect(result[0].replyToId).toBe("42");
+    expect(result[0].replyToTag).toBe(true);
+  });
 });

src/auto-reply/reply/reply-payloads.ts

@@ -62,10 +62,15 @@ export function applyReplyThreading(params: {
 }): ReplyPayload[] {
   const { payloads, replyToMode, replyToChannel, currentMessageId } = params;
   const applyReplyToMode = createReplyToModeFilterForChannel(replyToMode, replyToChannel);
+  const implicitReplyToId = currentMessageId?.trim() || undefined;
   return payloads
-    .map((payload) =>
+    .map((payload) => {
-      applyReplyTagsToPayload({ ...payload, replyToCurrent: true }, currentMessageId),
+      const autoThreaded =
-    )
+        payload.replyToId || payload.replyToCurrent === false || !implicitReplyToId
+          ? payload
+          : { ...payload, replyToId: implicitReplyToId };
+      return applyReplyTagsToPayload(autoThreaded, currentMessageId);
+    })
     .filter(isRenderablePayload)
     .map(applyReplyToMode);
 }