← Back to Skills Marketplace
40
Downloads
0
Stars
0
Active Installs
1
Versions
Install in OpenClaw
/install feishu-tts-pro
Description
将文字转换为语音并以飞书语音气泡消息发送。使用 Edge-TTS（微软，免费，无限次）生成中文语音，FFmpeg 转码为 ogg/opus，上传到飞书作为 audio 类型消息。支持自定义音色、接收者和 Python 环境。当用户要求用语音回复、发语音消息、TTS 朗读内容时触发。
README (SKILL.md)
feishu-tts-pro
将中文文字转换为语音，直接发送为飞书语音气泡消息。
使用方法
# 发送语音消息（默认发送给 FEISHU_DEFAULT_USER）
tts-send "你好，这是番茄的语音回复"
# 发送给指定用户
tts-send "你好" ou_xxxxx
环境变量（必填）
| 变量 | 说明 |
|---|---|
FEISHU_APP_ID |
飞书应用 App ID |
FEISHU_APP_SECRET |
飞书应用 App Secret |
FEISHU_DEFAULT_USER |
默认接收者的 open_id（可选，未提供时需要命令行参数） |
环境变量（可选）
| 变量 | 默认值 | 说明 |
|---|---|---|
EDGE_TTS_VOICE |
zh-CN-YunxiNeural |
Edge-TTS 音色 |
PYTHON_BIN |
系统 Python | Python 路径（用于 edge-tts） |
可选音色
| 音色 | 说明 |
|---|---|
zh-CN-YunxiNeural |
云希（男声，活泼阳光，默认） |
zh-CN-XiaoxiaoNeural |
晓晓（女声，温和） |
zh-CN-YunyangNeural |
云扬（男声，专业） |
zh-CN-XiaoyiNeural |
晓逸（女声，甜美） |
zh-CN-tianxiangNeural |
天祥（男声，大气） |
技术流程
- Edge-TTS 生成 MP3（使用指定音色）
- FFmpeg 转码为 ogg/opus（16kHz, mono, 飞书兼容）
- 上传到飞书（file_type=opus，duration 嵌入）
- 发送 audio 消息气泡
依赖
- Node.js 18+
- Python 3.8+ + edge-tts（
pip install edge-tts） - FFmpeg（系统安装）
安装 edge-tts
pip install edge-tts
或使用 uv：
uv pip install edge-tts --python /path/to/your/python
配置示例
在 AGENTS.md 或 openclaw.json 中配置：
{
"skills": {
"entries": {
"feishu-tts-pro": {
"env": {
"FEISHU_APP_ID": "cli_xxxxxxxxxxxx",
"FEISHU_APP_SECRET": "xxxxxxxxxxxxxxxx",
"FEISHU_DEFAULT_USER": "ou_xxxxxxxxxxxx",
"EDGE_TTS_VOICE": "zh-CN-YunxiNeural"
}
}
}
}
}
故障排除
TTS 生成失败
- 确认 edge-tts 已安装：
python -c "import edge_tts" - 确认 Python 路径正确
语音气泡无时长显示
- 上传时需要传入 duration 参数（毫秒）
- 脚本已自动处理
发送失败 (code=99991663)
- 检查 App ID/Secret 是否正确
- 确认飞书应用已开通「发消息」权限
Powered by 番茄 (OpenClaw Agent)
Usage Guidance
This skill's code appears to do what it claims, but there are small incoherences you should address before installing: 1) The registry marks EDGE_TTS_VOICE, PYTHON_BIN, and FEISHU_DEFAULT_USER as required env vars even though the README and script treat them as optional — don't provide more secrets than necessary; only FEISHU_APP_ID and FEISHU_APP_SECRET are essential. 2) The script calls ffprobe to measure duration but the metadata only requires ffmpeg; ensure ffprobe is available or add it to the requirements. 3) Review the FEISHU_APP_ID/FEISHU_APP_SECRET usage: installing this skill gives it the ability to obtain tenant tokens and send messages as the app — only grant an app with limited messaging scope and test in a controlled account. 4) Verify you have a safe Python environment with edge-tts installed (pip install edge-tts) and run the script in an isolated environment if you are uncertain. If you want higher assurance, ask the publisher for corrected metadata (declare ffprobe and mark optional env vars as optional) or request a signed / official source URL for the skill.
Capability Analysis
Type: OpenClaw Skill
Name: feishu-tts-pro
Version: 1.0.0
The feishu-tts-pro skill is a legitimate tool for converting text to speech and sending it via Feishu. It uses Node.js to orchestrate Python (edge-tts) and FFmpeg for audio processing, with all network requests directed to official Feishu API endpoints (open.feishu.cn). The script bin/tts-send.js handles inputs safely using JSON stringification when spawning sub-processes, and no indicators of data exfiltration or malicious intent were found.
Capability Assessment
Purpose & Capability
Name/description, SKILL.md, and the included bin/tts-send.js are consistent: they generate TTS via edge-tts, transcode with FFmpeg, and upload/send audio to Feishu. Requesting FEISHU_APP_ID and FEISHU_APP_SECRET is appropriate for Feishu integration.
Instruction Scope
Runtime instructions and the script stay within the stated scope: they spawn Python to run edge-tts, call ffmpeg/ffprobe for audio processing, write temporary files to the OS temp dir, and call only open.feishu.cn endpoints to upload and send messages. There are no unexpected external endpoints or broad file reads.
Install Mechanism
No install spec (instruction-only with a helper script). The SKILL.md documents installing edge-tts via pip; nothing is downloaded from arbitrary URLs. This is low install risk.
Credentials
Registry metadata lists FEISHU_APP_ID, FEISHU_APP_SECRET, FEISHU_DEFAULT_USER, EDGE_TTS_VOICE, and PYTHON_BIN as required env vars, but SKILL.md and the script treat FEISHU_DEFAULT_USER, EDGE_TTS_VOICE, and PYTHON_BIN as optional (have defaults). The mismatch is confusing and could cause over-provisioning of secrets. The only truly necessary secrets for Feishu are the App ID/Secret.
Persistence & Privilege
always is false and the skill does not request system-wide configuration or permanently modify other skills. The skill uses temporary files and obtains transient tenant_access_token from Feishu — expected behavior for this function.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install feishu-tts-pro - After installation, invoke the skill by name or use
/feishu-tts-pro - Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.0
- Initial release of feishu-tts-pro.
- Converts Chinese text to speech and sends as Feishu audio bubble messages.
- Uses Edge-TTS for TTS generation, FFmpeg for audio conversion to ogg/opus.
- Supports custom voice, recipient selection, and Python environment configuration.
- Requires environment variables for Feishu API and voice settings.
- Includes usage guidance, troubleshooting tips, and installation instructions.
Metadata
Frequently Asked Questions
What is 飞书TTS Pro?
将文字转换为语音并以飞书语音气泡消息发送。使用 Edge-TTS（微软，免费，无限次）生成中文语音，FFmpeg 转码为 ogg/opus，上传到飞书作为 audio 类型消息。支持自定义音色、接收者和 Python 环境。当用户要求用语音回复、发语音消息、TTS 朗读内容时触发。 It is an AI Agent Skill for Claude Code / OpenClaw, with 40 downloads so far.
How do I install 飞书TTS Pro?
Run "/install feishu-tts-pro" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is 飞书TTS Pro free?
Yes, 飞书TTS Pro is completely free, licensed under MIT-0. You can download, install and use it at no cost.
Which platforms does 飞书TTS Pro support?
飞书TTS Pro is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created 飞书TTS Pro?
It is built and maintained by minybear (@minybear); the current version is v1.0.0.
More Skills