url2podcast

Use when 需要把网页内容转成双人中文播客，并按脚本逐句参数合成音频后输出成品

This skill appears to do what it says: it fetches a webpage, generates a two-person podcast script, and calls ByteDance's (Volc) TTS API to synthesize audio, then merges audio with ffmpeg. Before installing/providing credentials, consider: 1) any webpage text sent to the TTS endpoint will leave your environment—do not send sensitive/private content to a third-party TTS service; 2) VOLC_APPID and VOLC_TOKEN are secrets—treat them like API keys and rotate/revoke if compromised; 3) the skill will clear and reuse skills/podcast-maker/workspace each run—ensure you have no other critical data there; 4) the Authorization header shown uses an unusual 'Bearer;${VOLC_TOKEN}' format (note the semicolon) — confirm with your Volc docs that this is correct. If you need stronger guarantees about data handling or offline synthesis, consider running local TTS or reviewing an implementation that avoids external APIs.

Type: OpenClaw Skill Name: url2podcast Version: 1.0.0 The skill is a legitimate tool designed to convert web content into a two-person podcast using the ByteDance Volcano Engine TTS API. It follows a structured workflow involving content extraction, script generation, and audio synthesis using standard utilities like curl, ffmpeg, and a small Python helper for base64 decoding. The instructions in SKILL.md include security-conscious constraints, such as restricting file operations to a specific workspace and clearing it before use, and there is no evidence of data exfiltration or malicious intent.