fix: process tts in cron announce delivery

steipete · steipete · commit 47c020bfc4c0 · 2026-05-02T02:10:59.000+01:00
diff --git a/CHANGELOG.md b/CHANGELOG.md
@@ -16,6 +16,7 @@ Docs: https://docs.openclaw.ai
 ### Fixes
 
 - Gateway/sessions: move hot transcript reads and mirror appends onto async bounded IO with serialized parent-linked writes, keeping large session histories from stalling Gateway requests and channel replies. Fixes #75656. Thanks @DerFlash.
+- Cron/TTS: run cron announce payloads through the normal TTS directive transform before outbound delivery, so scheduled `[[tts]]` replies generate voice payloads instead of leaking raw tags. Fixes #52125. Thanks @kenchen3000.
 - Doctor/WhatsApp: warn when Linux crontabs still run the legacy `ensure-whatsapp.sh` health check, which can misreport `Gateway inactive` when cron lacks the systemd user-bus environment. Fixes #60204. Thanks @mySebbe.
 - Slack/setup: print the generated app manifest as plain JSON instead of embedding it inside the framed setup note, so it can be copied into Slack without deleting border characters. Fixes #65751. Thanks @theDanielJLewis.
 - Channels/WhatsApp: route CLI logout through the live Gateway and stop runtime-backed listeners before channel removal, so removing a WhatsApp account does not leave the old socket replying until restart. Fixes #67746. Thanks @123Mismail.
diff --git a/src/cron/isolated-agent/delivery-dispatch.double-announce.test.ts b/src/cron/isolated-agent/delivery-dispatch.double-announce.test.ts
@@ -15,10 +15,12 @@ import { SILENT_REPLY_TOKEN } from "../../auto-reply/tokens.js";
 
 // --- Module mocks (must be hoisted before imports) ---
 
-const { countActiveDescendantRunsMock, retireSessionMcpRuntimeMock } = vi.hoisted(() => ({
-  countActiveDescendantRunsMock: vi.fn().mockReturnValue(0),
-  retireSessionMcpRuntimeMock: vi.fn().mockResolvedValue(true),
-}));
+const { countActiveDescendantRunsMock, maybeApplyTtsToPayloadMock, retireSessionMcpRuntimeMock } =
+  vi.hoisted(() => ({
+    countActiveDescendantRunsMock: vi.fn().mockReturnValue(0),
+    maybeApplyTtsToPayloadMock: vi.fn(async (params: { payload: unknown }) => params.payload),
+    retireSessionMcpRuntimeMock: vi.fn().mockResolvedValue(true),
+  }));
 
 vi.mock("../../config/sessions/main-session.js", () => ({
   resolveAgentMainSessionKey: vi.fn(({ agentId }: { agentId: string }) => `agent:${agentId}:main`),
@@ -66,6 +68,10 @@ vi.mock("../../infra/system-events.js", () => ({
   enqueueSystemEvent: vi.fn(),
 }));
 
+vi.mock("../../tts/tts.runtime.js", () => ({
+  maybeApplyTtsToPayload: maybeApplyTtsToPayloadMock,
+}));
+
 vi.mock("./subagent-followup-hints.js", () => ({
   expectsSubagentFollowup: vi.fn().mockReturnValue(false),
   isLikelyInterimCronMessage: vi.fn().mockReturnValue(false),
@@ -181,6 +187,7 @@ describe("dispatchCronDelivery — double-announce guard", () => {
     vi.mocked(readDescendantSubagentFallbackReply).mockResolvedValue(undefined);
     vi.mocked(waitForDescendantSubagentSummary).mockResolvedValue(undefined);
     vi.mocked(retireSessionMcpRuntime).mockResolvedValue(true);
+    maybeApplyTtsToPayloadMock.mockReset().mockImplementation(async (params) => params.payload);
   });
 
   afterEach(() => {
@@ -336,6 +343,62 @@ describe("dispatchCronDelivery — double-announce guard", () => {
     ).toBe(false);
   });
 
+  it("applies TTS directives before direct cron announce delivery", async () => {
+    vi.mocked(countActiveDescendantRuns).mockReturnValue(0);
+    vi.mocked(isLikelyInterimCronMessage).mockReturnValue(false);
+    maybeApplyTtsToPayloadMock.mockImplementation(async (params: { payload: unknown }) => {
+      const payload = params.payload as { text?: string };
+      expect(payload.text).toBe("[[tts]] Morning briefing complete.");
+      return {
+        text: "Morning briefing complete.",
+        mediaUrl: "file:///tmp/cron-tts.mp3",
+        audioAsVoice: true,
+        spokenText: "Morning briefing complete.",
+      };
+    });
+
+    const params = makeBaseParams({
+      synthesizedText: "[[tts]] Morning briefing complete.",
+      runStartedAt: 1_000,
+    });
+    params.cfgWithAgentDefaults = {
+      messages: {
+        tts: {
+          auto: "tagged",
+          provider: "microsoft",
+        },
+      },
+    } as never;
+
+    const state = await dispatchCronDelivery(params);
+
+    expect(state.deliveryAttempted).toBe(true);
+    expect(state.delivered).toBe(true);
+    expect(maybeApplyTtsToPayloadMock).toHaveBeenCalledWith(
+      expect.objectContaining({
+        cfg: params.cfgWithAgentDefaults,
+        channel: "telegram",
+        kind: "final",
+        agentId: "main",
+        accountId: undefined,
+      }),
+    );
+    expect(deliverOutboundPayloads).toHaveBeenCalledWith(
+      expect.objectContaining({
+        channel: "telegram",
+        to: "123456",
+        payloads: [
+          {
+            text: "Morning briefing complete.",
+            mediaUrl: "file:///tmp/cron-tts.mp3",
+            audioAsVoice: true,
+            spokenText: "Morning briefing complete.",
+          },
+        ],
+      }),
+    );
+  });
+
   it("preserves all successful text payloads for direct delivery", async () => {
     vi.mocked(countActiveDescendantRuns).mockReturnValue(0);
     vi.mocked(isLikelyInterimCronMessage).mockReturnValue(false);
diff --git a/src/cron/isolated-agent/delivery-dispatch.ts b/src/cron/isolated-agent/delivery-dispatch.ts
@@ -13,6 +13,7 @@ import {
   resolveMainSessionKey,
 } from "../../config/sessions/main-session.js";
 import type { OpenClawConfig } from "../../config/types.openclaw.js";
+import type { TtsAutoMode } from "../../config/types.tts.js";
 import { sleepWithAbort } from "../../infra/backoff.js";
 import { formatErrorMessage } from "../../infra/errors.js";
 import type { OutboundDeliveryResult } from "../../infra/outbound/deliver.js";
@@ -24,6 +25,7 @@ import {
   normalizeOptionalLowercaseString,
   normalizeOptionalString,
 } from "../../shared/string-coerce.js";
+import { shouldAttemptTtsPayload } from "../../tts/tts-config.js";
 import { createCronExecutionId } from "../run-id.js";
 import { hasScheduledNextRunAtMs } from "../service/jobs.js";
 import type { CronJob, CronRunTelemetry } from "../types.js";
@@ -119,6 +121,7 @@ type DispatchCronDeliveryParams = {
   deliveryPayloadHasStructuredContent: boolean;
   deliveryPayloads: ReplyPayload[];
   synthesizedText?: string;
+  ttsAuto?: TtsAutoMode;
   summary?: string;
   outputText?: string;
   telemetry?: CronRunTelemetry;
@@ -183,6 +186,7 @@ let deliveryLoggerRuntimePromise:
 let subagentFollowupRuntimePromise:
   | Promise<typeof import("./subagent-followup.runtime.js")>
   | undefined;
+let ttsRuntimePromise: Promise<typeof import("../../tts/tts.runtime.js")> | undefined;
 
 const COMPLETED_DIRECT_CRON_DELIVERIES = new Map<string, CompletedDirectCronDelivery>();
 
@@ -217,6 +221,11 @@ async function loadSubagentFollowupRuntime(): Promise<
   return await subagentFollowupRuntimePromise;
 }
 
+async function loadTtsRuntime(): Promise<typeof import("../../tts/tts.runtime.js")> {
+  ttsRuntimePromise ??= import("../../tts/tts.runtime.js");
+  return await ttsRuntimePromise;
+}
+
 async function logCronDeliveryWarn(message: string): Promise<void> {
   const { logWarn } = await loadDeliveryLoggerRuntime();
   logWarn(message);
@@ -303,6 +312,40 @@ function getCompletedDirectCronDelivery(
   return cloneDeliveryResults(cached.results);
 }
 
+async function maybeApplyTtsToCronPayloads(params: {
+  cfg: OpenClawConfig;
+  payloads: ReplyPayload[];
+  delivery: SuccessfulDeliveryTarget;
+  agentId: string;
+  ttsAuto?: TtsAutoMode;
+}): Promise<ReplyPayload[]> {
+  if (
+    !shouldAttemptTtsPayload({
+      cfg: params.cfg,
+      ttsAuto: params.ttsAuto,
+      agentId: params.agentId,
+      channelId: params.delivery.channel,
+      accountId: params.delivery.accountId,
+    })
+  ) {
+    return params.payloads;
+  }
+  const { maybeApplyTtsToPayload } = await loadTtsRuntime();
+  return await Promise.all(
+    params.payloads.map((payload) =>
+      maybeApplyTtsToPayload({
+        payload,
+        cfg: params.cfg,
+        channel: params.delivery.channel,
+        kind: "final",
+        ttsAuto: params.ttsAuto,
+        agentId: params.agentId,
+        accountId: params.delivery.accountId,
+      }),
+    ),
+  );
+}
+
 function buildDirectCronDeliveryIdempotencyKey(params: {
   jobId: string;
   runStartedAt: number;
@@ -524,7 +567,7 @@ export async function dispatchCronDelivery(
           : synthesizedText
             ? [{ text: synthesizedText }]
             : [];
-      const payloadsForDelivery = rawPayloads
+      const normalizedPayloads = rawPayloads
         .map((p) => {
           if (!p.text) {
             return p;
@@ -535,7 +578,7 @@ export async function dispatchCronDelivery(
           });
         })
         .filter((p) => hasReplyPayloadContent(p, { trimText: true }));
-      if (payloadsForDelivery.length === 0) {
+      if (normalizedPayloads.length === 0) {
         return await finishSilentReplyDelivery();
       }
       if (params.isAborted()) {
@@ -575,6 +618,18 @@ export async function dispatchCronDelivery(
           ...params.telemetry,
         });
       }
+      const payloadsForDelivery = (
+        await maybeApplyTtsToCronPayloads({
+          cfg: params.cfgWithAgentDefaults,
+          payloads: normalizedPayloads,
+          delivery,
+          agentId: params.agentId,
+          ttsAuto: params.ttsAuto,
+        })
+      ).filter((p) => hasReplyPayloadContent(p, { trimText: true }));
+      if (payloadsForDelivery.length === 0) {
+        return await finishSilentReplyDelivery();
+      }
       deliveryAttempted = true;
       const cachedResults = getCompletedDirectCronDelivery(deliveryIdempotencyKey);
       if (cachedResults) {
diff --git a/src/cron/isolated-agent/run.ts b/src/cron/isolated-agent/run.ts
@@ -958,6 +958,7 @@ async function finalizeCronRun(params: {
     deliveryPayloadHasStructuredContent,
     deliveryPayloads,
     synthesizedText,
+    ttsAuto: prepared.cronSession.sessionEntry.ttsAuto,
     summary,
     outputText,
     telemetry,
