feishu-audio-msg_skill

Overview

This skill sends text-to-speech audio into Feishu chats as a native playable audio message (msg_type=audio), not as a file attachment. It automates the Feishu two-step flow so TTS or voice announcements appear as proper voice messages in group chats or DMs. Use it when you need speech announcements, voice roll calls, or any TTS delivered directly inside Feishu conversations.

How this skill works

The skill performs a two-step Feishu API sequence: first it uploads the audio with file_type=opus to obtain a file_key, then it posts a message with msg_type=audio and the returned file_key so the audio is shown as a playable message. It reads credentials automatically from standard OpenClaw locations when available, or you can supply a Feishu token for custom integrations. The script accepts text and target receive_id (chat_id or open_id) and supports selecting different TTS voices.

When to use it

• Send TTS or speech announcements to a Feishu group or user
• Deliver voice roll calls (语音报数) or automated spoken reminders
• Convert short text outputs into playable Feishu voice messages
• Replace attachments with native Feishu audio playback for better UX
• Integrate speech output into chatbots that post directly to Feishu

Best practices

• Always upload with file_type=opus, even if your source is MP3 or WAV
• Use concise TTS segments (short clips) for reliable playback and speed
• Provide receive_id as chat_id (oc_xxx for groups) or open_id (ou_xxx for DMs)
• Prefer server-side token management; store Feishu creds in OpenClaw config
• Check API responses for upload file_key before sending the audio message

Example use cases

• Notify a project group in Feishu with a spoken status summary
• Run an automated voice roll call in a classroom or team meeting chat
• Send daily spoken KPIs to a manager DM using English or Chinese TTS
• Trigger short audible alerts from monitoring systems into a Feishu channel
• Convert chatbot replies to voice and post them as playable messages