feishu-edge-tts-win

飞书语音消息发送技能（Windows 版）。使用 Edge TTS（微软，免费）生成语音并以飞书语音气泡发送。

Security Scan

The skill's code matches its stated purpose (generate Edge TTS and send a Feishu audio message), but the package metadata omits important resource/credential requirements (it reads your openclaw.json for Feishu appId/appSecret and relies on ffmpeg/curl/edge-tts) — this mismatch warrants caution before installing or running.

Purpose & Capability

The name/description (Feishu TTS via Edge TTS on Windows) aligns with the included script: it generates MP3 via edge-tts, converts to OPUS with ffmpeg, uploads to Feishu, and sends audio. However the registry metadata claims no required config or credentials even though the script requires Feishu appId/appSecret from ~/.openclaw/openclaw.json.

Instruction Scope

SKILL.md and send_voice.py keep to the stated task: they generate audio, convert format, upload to Feishu API, and send an audio message. The instructions explicitly require the user to pass a config path and mention ffmpeg and edge-tts. The script does not read unrelated files or environment variables beyond the provided openclaw.json.

Install Mechanism

This is instruction-only with no install spec (low disk/write risk). The README asks users to pip install edge-tts and requires external ffmpeg and curl binaries — these are not enforced by the registry but are necessary at runtime.

Credentials

The script reads the user's openclaw.json to obtain Feishu appId/appSecret (sensitive credentials). The skill metadata did not declare required config paths or credentials. Requesting and using these credentials is reasonable for the feature, but the omission in metadata is a risk/opacity issue and users should ensure the config file only contains credentials they trust the skill to use.

Persistence & Privilege

always is false; the skill does not request persistent platform privileges or modify other skills. It runs as a one-shot script and does not attempt to persist or alter global agent settings.

What to consider before installing

This skill's code implements exactly what it claims: converting text to speech (via edge-tts), converting to OPUS (ffmpeg), uploading to Feishu, and sending an audio message. Before running: 1) Inspect your ~/.openclaw/openclaw.json and be aware it will be read for channels.feishu.appId and appSecret — only use credentials you are willing to expose to this script. 2) Ensure ffmpeg and curl are installed and on PATH and that you have installed the edge-tts Python package (the registry metadata did not list these requirements). 3) Consider testing with a throwaway Feishu app/account to verify behavior. 4) Because the script makes network calls (Microsoft Edge TTS endpoints via edge-tts and Feishu open APIs), do not run it with sensitive proprietary text you don't want transmitted externally. The primary concern is the metadata omission about required binaries and the fact the script reads your stored Feishu credentials; the code itself appears coherent and not intentionally malicious, but exercise caution and verify credentials and environment before use.

飞书语音消息发送（Windows）

快速使用

在 skills\feishu-edge-tts-win\scripts 目录执行：

python .\send_voice.py "要发送的文本" YOUR_FEISHU_OPEN_ID --voice zh-CN-XiaoxiaoNeural --config %USERPROFILE%\.openclaw\openclaw.json

也可以从任意目录执行：

python %USERPROFILE%\.openclaw\workspace\skills\feishu-edge-tts-win\scripts\send_voice.py "要发送的文本" YOUR_FEISHU_OPEN_ID --voice zh-CN-XiaoxiaoNeural --config %USERPROFILE%\.openclaw\openclaw.json

工作流程

1. edge-tts 生成 MP3
2. ffmpeg 转换为 OPUS（飞书语音格式）
3. 飞书 API 上传 OPUS 获取 file_key
4. 发送 msg_type=audio 消息

关键点：必须使用 msg_type: audio + file_key，否则会显示为普通文件而不是语音气泡。

音色

仅使用：zh-CN-XiaoxiaoNeural

依赖安装

pip install edge-tts

ffmpeg 需要可执行（已加入 PATH）。可用 ffmpeg -version 自检。

飞书配置

脚本读取 openclaw.json 的 channels.feishu.appId 和 channels.feishu.appSecret。 请在命令中显式传入 --config，例如：

--config %USERPROFILE%\.openclaw\openclaw.json