Skip to content

Agent Zero AI framework

Send Telegram media attachments natively

Anmol Malik97ef4797e0ec
diff --git a/plugins/_telegram_integration/helpers/handler.py b/plugins/_telegram_integration/helpers/handler.py
index 07c241478..3fd0f8c82 100644
--- a/plugins/_telegram_integration/helpers/handler.py
+++ b/plugins/_telegram_integration/helpers/handler.py
@@ -526,6 +526,12 @@ async def send_telegram_reply(
local_path = files.fix_dev_path(path)
if tc.is_image_file(local_path):
await tc.send_photo(reply_bot, chat_id, local_path, reply_to_message_id=reply_to)
+ elif tc.is_voice_file(local_path):
+ await tc.send_voice(reply_bot, chat_id, local_path, reply_to_message_id=reply_to)
+ elif tc.is_audio_file(local_path):
+ await tc.send_audio(reply_bot, chat_id, local_path, reply_to_message_id=reply_to)
+ elif tc.is_video_file(local_path):
+ await tc.send_video(reply_bot, chat_id, local_path, reply_to_message_id=reply_to)
else:
await tc.send_file(reply_bot, chat_id, local_path, reply_to_message_id=reply_to)
diff --git a/plugins/_telegram_integration/helpers/telegram_client.py b/plugins/_telegram_integration/helpers/telegram_client.py
index 45db75977..2fc9f69f1 100644
--- a/plugins/_telegram_integration/helpers/telegram_client.py
+++ b/plugins/_telegram_integration/helpers/telegram_client.py
@@ -115,6 +115,81 @@ async def send_photo(
return None
+async def send_voice(
+ bot: Bot,
+ chat_id: int,
+ voice_path: str,
+ caption: str = "",
+ reply_to_message_id: int | None = None,
+) -> int | None:
+ """Send a voice message from local path. Returns message_id or None on error."""
+ try:
+ if not os.path.isfile(voice_path):
+ PrintStyle.error(f"Telegram: voice file not found: {voice_path}")
+ return None
+ input_file = FSInputFile(voice_path)
+ msg = await bot.send_voice(
+ chat_id=chat_id,
+ voice=input_file,
+ caption=caption[:1024] if caption else None,
+ reply_to_message_id=reply_to_message_id,
+ )
+ return msg.message_id
+ except Exception as e:
+ PrintStyle.error(f"Telegram send_voice failed: {format_error(e)}")
+ return None
+
+
+async def send_audio(
+ bot: Bot,
+ chat_id: int,
+ audio_path: str,
+ caption: str = "",
+ reply_to_message_id: int | None = None,
+) -> int | None:
+ """Send an audio message from local path. Returns message_id or None on error."""
+ try:
+ if not os.path.isfile(audio_path):
+ PrintStyle.error(f"Telegram: audio file not found: {audio_path}")
+ return None
+ input_file = FSInputFile(audio_path)
+ msg = await bot.send_audio(
+ chat_id=chat_id,
+ audio=input_file,
+ caption=caption[:1024] if caption else None,
+ reply_to_message_id=reply_to_message_id,
+ )
+ return msg.message_id
+ except Exception as e:
+ PrintStyle.error(f"Telegram send_audio failed: {format_error(e)}")
+ return None
+
+
+async def send_video(
+ bot: Bot,
+ chat_id: int,
+ video_path: str,
+ caption: str = "",
+ reply_to_message_id: int | None = None,
+) -> int | None:
+ """Send a video message from local path. Returns message_id or None on error."""
+ try:
+ if not os.path.isfile(video_path):
+ PrintStyle.error(f"Telegram: video file not found: {video_path}")
+ return None
+ input_file = FSInputFile(video_path)
+ msg = await bot.send_video(
+ chat_id=chat_id,
+ video=input_file,
+ caption=caption[:1024] if caption else None,
+ reply_to_message_id=reply_to_message_id,
+ )
+ return msg.message_id
+ except Exception as e:
+ PrintStyle.error(f"Telegram send_video failed: {format_error(e)}")
+ return None
+
+
# Inline keyboards
def build_inline_keyboard(
@@ -294,6 +369,9 @@ def _split_text(text: str, max_len: int) -> list[str]:
_IMAGE_EXTENSIONS = {".jpg", ".jpeg", ".png", ".gif", ".webp", ".bmp"}
+_VOICE_EXTENSIONS = {".ogg", ".oga", ".opus"}
+_AUDIO_EXTENSIONS = {".mp3", ".m4a", ".aac", ".wav", ".flac"}
+_VIDEO_EXTENSIONS = {".mp4", ".m4v", ".mov", ".webm"}
def is_image_file(path: str) -> bool:
@@ -301,6 +379,21 @@ def is_image_file(path: str) -> bool:
return ext in _IMAGE_EXTENSIONS
+def is_voice_file(path: str) -> bool:
+ _, ext = os.path.splitext(path.lower())
+ return ext in _VOICE_EXTENSIONS
+
+
+def is_audio_file(path: str) -> bool:
+ _, ext = os.path.splitext(path.lower())
+ return ext in _AUDIO_EXTENSIONS
+
+
+def is_video_file(path: str) -> bool:
+ _, ext = os.path.splitext(path.lower())
+ return ext in _VIDEO_EXTENSIONS
+
+
def md_to_telegram_html(text: str) -> str:
"""Convert Markdown to Telegram-compatible HTML."""
stash: list[str] = []
diff --git a/tests/test_telegram_media_delivery.py b/tests/test_telegram_media_delivery.py
new file mode 100644
index 000000000..bbdb2a2bc
--- /dev/null
+++ b/tests/test_telegram_media_delivery.py
@@ -0,0 +1,108 @@
+import asyncio
+
+from plugins._telegram_integration.helpers import handler
+from plugins._telegram_integration.helpers import telegram_client as tc
+from plugins._telegram_integration.helpers.constants import (
+ CTX_TG_BOT,
+ CTX_TG_CHAT_ID,
+ CTX_TG_REPLY_TO,
+)
+
+
+class FakeBotInstance:
+ class Bot:
+ token = "token"
+
+ bot = Bot()
+
+
+class FakeContext:
+ def __init__(self):
+ self.data = {
+ CTX_TG_BOT: "main",
+ CTX_TG_CHAT_ID: 123,
+ CTX_TG_REPLY_TO: 456,
+ }
+
+
+class FakeTempBot:
+ async def __aenter__(self):
+ return object()
+
+ async def __aexit__(self, exc_type, exc, tb):
+ return False
+
+
+class FakeMediaBot:
+ def __init__(self):
+ self.calls = []
+
+ async def send_voice(self, **kwargs):
+ self.calls.append(("voice", kwargs))
+ return type("Message", (), {"message_id": 11})()
+
+ async def send_audio(self, **kwargs):
+ self.calls.append(("audio", kwargs))
+ return type("Message", (), {"message_id": 12})()
+
+ async def send_video(self, **kwargs):
+ self.calls.append(("video", kwargs))
+ return type("Message", (), {"message_id": 13})()
+
+
+def test_telegram_client_native_media_helpers(monkeypatch):
+ bot = FakeMediaBot()
+ monkeypatch.setattr(tc.os.path, "isfile", lambda path: True)
+
+ voice_id = asyncio.run(tc.send_voice(bot, 123, "voice.ogg", reply_to_message_id=456))
+ audio_id = asyncio.run(tc.send_audio(bot, 123, "song.mp3", reply_to_message_id=456))
+ video_id = asyncio.run(tc.send_video(bot, 123, "clip.mp4", reply_to_message_id=456))
+
+ assert (voice_id, audio_id, video_id) == (11, 12, 13)
+ assert [kind for kind, _ in bot.calls] == ["voice", "audio", "video"]
+ assert bot.calls[0][1]["chat_id"] == 123
+ assert bot.calls[0][1]["reply_to_message_id"] == 456
+
+
+def test_telegram_reply_routes_native_media_attachments(monkeypatch):
+ calls = []
+
+ def fake_temp_bot(*args, **kwargs):
+ return FakeTempBot()
+
+ async def record(kind, bot, chat_id, path, reply_to_message_id=None, **kwargs):
+ calls.append(
+ {
+ "kind": kind,
+ "chat_id": chat_id,
+ "path": path,
+ "reply_to_message_id": reply_to_message_id,
+ }
+ )
+ return len(calls)
+
+ monkeypatch.setattr(handler, "get_bot", lambda name: FakeBotInstance())
+ monkeypatch.setattr(handler, "_temp_bot", fake_temp_bot)
+ monkeypatch.setattr(handler.files, "fix_dev_path", lambda path: path)
+ monkeypatch.setattr(handler.tc, "send_photo", lambda *a, **kw: record("photo", *a, **kw))
+ monkeypatch.setattr(handler.tc, "send_voice", lambda *a, **kw: record("voice", *a, **kw))
+ monkeypatch.setattr(handler.tc, "send_audio", lambda *a, **kw: record("audio", *a, **kw))
+ monkeypatch.setattr(handler.tc, "send_video", lambda *a, **kw: record("video", *a, **kw))
+ monkeypatch.setattr(handler.tc, "send_file", lambda *a, **kw: record("document", *a, **kw))
+
+ error = asyncio.run(
+ handler.send_telegram_reply(
+ FakeContext(),
+ "",
+ ["image.png", "voice.ogg", "song.mp3", "clip.mp4", "archive.zip"],
+ )
+ )
+
+ assert error is None
+ assert calls == [
+ {"kind": "photo", "chat_id": 123, "path": "image.png", "reply_to_message_id": 456},
+ {"kind": "voice", "chat_id": 123, "path": "voice.ogg", "reply_to_message_id": 456},
+ {"kind": "audio", "chat_id": 123, "path": "song.mp3", "reply_to_message_id": 456},
+ {"kind": "video", "chat_id": 123, "path": "clip.mp4", "reply_to_message_id": 456},
+ {"kind": "document", "chat_id": 123, "path": "archive.zip", "reply_to_message_id": 456},
+ ]