fix(agents): keep voice media with no-reply sentinel

CHANGELOG.md

@@ -76,6 +76,7 @@ Docs: https://docs.openclaw.ai
 - Exec approvals: allow bare command-name allowlist patterns to match PATH-resolved executable basenames without trusting `./tool` or absolute path-selected binaries. Fixes #71315. Thanks @chen-zhang-cs-code and @dengluozhang.
 - Config/recovery: skip whole-file last-known-good rollback when invalidity is scoped to `plugins.entries.*`, preserving unrelated user settings during plugin schema or host-version skew. Fixes #71289. Thanks @jalehman.
 - Agents/tools: keep resolved reply-run configs from being overwritten by stale runtime snapshots, and let empty web runtime metadata fall back to configured provider auto-detection so standard and queued turns expose the same tool set. Fixes #71355. Thanks @c-g14.
+- Agents/TTS: preserve voice media when a tool-generated reply is paired with an exact `NO_REPLY` sentinel, stripping the sentinel text instead of dropping the audio payload. Fixes #66092.
 - Compaction: honor explicit `agents.defaults.compaction.keepRecentTokens` for manual `/compact`, re-distill safeguard summaries instead of snowballing previous summaries, and enable safeguard summary quality checks by default. Fixes #71357. Thanks @WhiteGiverMa.
 - Sessions: honor configured `session.maintenance` settings during load-time maintenance instead of falling back to default entry caps. Fixes #71356. Thanks @comolago.
 - Browser/sandbox: pass the resolved `browser.ssrfPolicy` into sandbox browser bridges and refresh cached bridges when the effective policy changes, so sandboxed browser navigation honors private-network opt-ins. Fixes #45153 and #57055. Thanks @jzakirov, @zuoanCo, and @kybrcore.

src/agents/pi-embedded-runner/run/payloads.test.ts

@@ -200,6 +200,20 @@ describe("buildEmbeddedRunPayloads tool-error warnings", () => {
     });
   });
 
+  it("strips NO_REPLY text but keeps voice media directives", () => {
+    const payloads = buildPayloads({
+      assistantTexts: ["NO_REPLY\nMEDIA:/tmp/openclaw/tts-a/voice-a.opus\n[[audio_as_voice]]"],
+    });
+
+    expect(payloads).toHaveLength(1);
+    expect(payloads[0]).toMatchObject({
+      mediaUrl: "/tmp/openclaw/tts-a/voice-a.opus",
+      mediaUrls: ["/tmp/openclaw/tts-a/voice-a.opus"],
+      audioAsVoice: true,
+    });
+    expect(payloads[0]?.text).toBeUndefined();
+  });
+
   it("preserves media directives when stored assistant text was reduced to visible text only", () => {
     const payloads = buildPayloads({
       assistantTexts: ["Attached image"],

src/agents/pi-embedded-runner/run/payloads.ts

@@ -381,16 +381,27 @@ export function buildEmbeddedRunPayloads(params: {
 
   const hasAudioAsVoiceTag = replyItems.some((item) => item.audioAsVoice);
   return replyItems
-    .map((item) => ({
-      text: normalizeOptionalString(item.text),
-      mediaUrls: item.media?.length ? item.media : undefined,
-      mediaUrl: item.media?.[0],
-      isError: item.isError,
-      replyToId: item.replyToId,
-      replyToTag: item.replyToTag,
-      replyToCurrent: item.replyToCurrent,
-      audioAsVoice: item.audioAsVoice || Boolean(hasAudioAsVoiceTag && item.media?.length),
-    }))
+    .map((item) => {
+      const payload = {
+        text: normalizeOptionalString(item.text),
+        mediaUrls: item.media?.length ? item.media : undefined,
+        mediaUrl: item.media?.[0],
+        isError: item.isError,
+        replyToId: item.replyToId,
+        replyToTag: item.replyToTag,
+        replyToCurrent: item.replyToCurrent,
+        audioAsVoice: item.audioAsVoice || Boolean(hasAudioAsVoiceTag && item.media?.length),
+      };
+      if (payload.text && isSilentReplyPayloadText(payload.text, SILENT_REPLY_TOKEN)) {
+        const silentText = payload.text;
+        payload.text = undefined;
+        if (hasOutboundReplyContent(payload)) {
+          return payload;
+        }
+        payload.text = silentText;
+      }
+      return payload;
+    })
     .filter((p) => {
       if (!hasOutboundReplyContent(p)) {
         return false;