Feishu Voice Clone TTS Skill

Convert text to speech using Volcengine TTS with preset or cloned voices and send audio messages to Feishu chats or groups.

Security Scan

The code implements exactly what the skill claims (Volcengine TTS -> convert -> upload to Feishu), but metadata omissions and a couple undocumented local/config access points create small coherence concerns you should confirm before installing.

Purpose & Capability

The script's behavior (call Volcengine TTS, convert with ffmpeg/ffprobe, upload to Feishu) matches the skill description. However the registry metadata lists no required environment variables or binaries while the SKILL.md and code do require FEISHU_APP_ID, FEISHU_APP_SECRET, FEISHU_CHAT_ID, VOLC_API_KEY, VOLC_VOICE_TYPE and the ffmpeg/ffprobe binaries — this metadata omission is inconsistent and worth noting.

Instruction Scope

Runtime instructions and code limit actions to: calling Volcengine TTS API, converting audio locally with ffmpeg/ffprobe, obtaining a Feishu tenant token, uploading the resulting audio file, and sending a message. The SKILL.md documents these steps. The code does optionally read a local config file (~/.volcengine_key) as an alternative API-key source — this is documented in README but is an extra local file access to be aware of.

Install Mechanism

There is no install script or external download; the skill is delivered as source files. That reduces supply-chain risk. It does require system binaries (ffmpeg, ffprobe) and Python 'requests' at runtime; those prerequisites are mentioned in README but were not reflected in the registry's required-binaries field.

Credentials

The environment variables requested by the script (Feishu App ID/Secret, Feishu chat id, Volcengine API key and voice id) are appropriate and necessary for the described functionality. The script also supports reading ~/.volcengine_key for the Volcengine key — this config path was not declared in the registry metadata and gives the skill access to a home-directory file, which you should verify you want it to read.

Persistence & Privilege

The skill is not force-enabled (always:false), does not persist or modify other skills or system-wide settings, and only runs on explicit invocation. It exchanges credentials with the Feishu/Volcengine APIs as expected and uses temporary files for audio data.

What to consider before installing

This skill's code is coherent with its description, but the package/registry metadata omits required environment variables and binaries. Before installing: (1) Verify you trust the source — the author field/homepage are empty. (2) Provide dedicated Feishu and Volcengine credentials with minimal permissions (use a test app/service account), and do not reuse high-privilege secrets. (3) Ensure ffmpeg/ffprobe are installed in a controlled environment. (4) Be aware the script will read ~/.volcengine_key if present; remove or inspect that file if you don't want it used. (5) Review the code yourself (it is short and readable) or run it in an isolated container/VM. If you proceed, avoid giving production credentials until you’re comfortable; rotate them if you tested with real secrets.

使用火山引擎 TTS 将文本转换为语音，并发送到飞书。

支持两种方式：

1. 使用火山引擎预置的声音
2. 使用你自己克隆的声音

火山引擎配置

方式一：使用预置声音（无需克隆）

https://console.volcengine.com/speech/new/experience/tts?projectName=default

方式二：使用你自己克隆的声音

1. 声音克隆

https://console.volcengine.com/speech/new/experience/clone?projectName=default

2. 语音合成

https://console.volcengine.com/speech/new/experience/tts?projectName=default

配置

需要配置以下环境变量：

• FEISHU_APP_ID - 飞书 App ID
• FEISHU_APP_SECRET - 飞书 App Secret
• FEISHU_CHAT_ID - 飞书聊天 ID (user:xxx 或 chat:xxx)
• VOLC_API_KEY - 火山引擎 API Key
• VOLC_VOICE_TYPE - 火山引擎音色 ID（预置声音或克隆声音均可）

使用

用户: 用语音说"你好"
OpenClaw: [调用 TTS] [发送语音消息到飞书]

命令

• python feishu_tts.py "文本" - 发送语音消息
• python feishu_tts.py -c "chat:oc_xxx" "文本" - 发送到指定群聊