fix: allow media-only sends

src/telegram/bot-message-context.ts

@@ -98,6 +98,18 @@ export const buildTelegramMessageContext = async ({
     }
   };
 
+  const sendRecordVoice = async () => {
+    try {
+      await bot.api.sendChatAction(
+        chatId,
+        "record_voice",
+        buildTypingThreadParams(resolvedThreadId),
+      );
+    } catch (err) {
+      logVerbose(`telegram record_voice cue failed for chat ${chatId}: ${String(err)}`);
+    }
+  };
+
   // DM access control (secure defaults): "pairing" (default) / "allowlist" / "open" / "disabled"
   if (!isGroup) {
     if (dmPolicy === "disabled") return null;
@@ -408,6 +420,7 @@ export const buildTelegramMessageContext = async ({
     route,
     skillFilter,
     sendTyping,
+    sendRecordVoice,
     ackReactionPromise,
     reactionApi,
     removeAckAfterReply,

src/telegram/bot-message-dispatch.ts

@@ -37,6 +37,7 @@ export const dispatchTelegramMessage = async ({
     route,
     skillFilter,
     sendTyping,
+    sendRecordVoice,
     ackReactionPromise,
     reactionApi,
     removeAckAfterReply,
@@ -144,6 +145,7 @@ export const dispatchTelegramMessage = async ({
           replyToMode,
           textLimit,
           messageThreadId: resolvedThreadId,
+          onVoiceRecording: sendRecordVoice,
         });
         didSendReply = true;
       },

src/telegram/bot/delivery.test.ts

@@ -0,0 +1,77 @@
+import { beforeEach, describe, expect, it, vi } from "vitest";
+
+import type { Bot } from "grammy";
+
+import { deliverReplies } from "./delivery.js";
+
+const loadWebMedia = vi.fn();
+
+vi.mock("../../web/media.js", () => ({
+  loadWebMedia: (...args: unknown[]) => loadWebMedia(...args),
+}));
+
+vi.mock("grammy", () => ({
+  InputFile: class {
+    constructor(
+      public buffer: Buffer,
+      public fileName?: string,
+    ) {}
+  },
+}));
+
+describe("deliverReplies", () => {
+  beforeEach(() => {
+    loadWebMedia.mockReset();
+  });
+
+  it("skips audioAsVoice-only payloads without logging an error", async () => {
+    const runtime = { error: vi.fn() };
+    const bot = { api: {} } as unknown as Bot;
+
+    await deliverReplies({
+      replies: [{ audioAsVoice: true }],
+      chatId: "123",
+      token: "tok",
+      runtime,
+      bot,
+      replyToMode: "off",
+      textLimit: 4000,
+    });
+
+    expect(runtime.error).not.toHaveBeenCalled();
+  });
+
+  it("invokes onVoiceRecording before sending a voice note", async () => {
+    const events: string[] = [];
+    const runtime = { error: vi.fn() };
+    const sendVoice = vi.fn(async () => {
+      events.push("sendVoice");
+      return { message_id: 1, chat: { id: "123" } };
+    });
+    const bot = { api: { sendVoice } } as unknown as Bot;
+    const onVoiceRecording = vi.fn(async () => {
+      events.push("recordVoice");
+    });
+
+    loadWebMedia.mockResolvedValueOnce({
+      buffer: Buffer.from("voice"),
+      contentType: "audio/ogg",
+      fileName: "note.ogg",
+    });
+
+    await deliverReplies({
+      replies: [{ mediaUrl: "https://example.com/note.ogg", audioAsVoice: true }],
+      chatId: "123",
+      token: "tok",
+      runtime,
+      bot,
+      replyToMode: "off",
+      textLimit: 4000,
+      onVoiceRecording,
+    });
+
+    expect(onVoiceRecording).toHaveBeenCalledTimes(1);
+    expect(sendVoice).toHaveBeenCalledTimes(1);
+    expect(events).toEqual(["recordVoice", "sendVoice"]);
+  });
+});

src/telegram/bot/delivery.ts

@@ -25,12 +25,19 @@ export async function deliverReplies(params: {
   replyToMode: ReplyToMode;
   textLimit: number;
   messageThreadId?: number;
+  /** Callback invoked before sending a voice message to switch typing indicator. */
+  onVoiceRecording?: () => Promise<void> | void;
 }) {
   const { replies, chatId, runtime, bot, replyToMode, textLimit, messageThreadId } = params;
   const threadParams = buildTelegramThreadParams(messageThreadId);
   let hasReplied = false;
   for (const reply of replies) {
-    if (!reply?.text && !reply?.mediaUrl && !(reply?.mediaUrls?.length ?? 0)) {
+    const hasMedia = Boolean(reply?.mediaUrl) || (reply?.mediaUrls?.length ?? 0) > 0;
+    if (!reply?.text && !hasMedia) {
+      if (reply?.audioAsVoice) {
+        logVerbose("telegram reply has audioAsVoice without media/text; skipping");
+        continue;
+      }
       runtime.error?.(danger("reply missing text/media"));
       continue;
     }
@@ -99,6 +106,8 @@ export async function deliverReplies(params: {
         });
         if (useVoice) {
           // Voice message - displays as round playable bubble (opt-in via [[audio_as_voice]])
+          // Switch typing indicator to record_voice before sending.
+          await params.onVoiceRecording?.();
           await bot.api.sendVoice(chatId, file, {
             ...mediaParams,
           });