Generate 1080x1920 vertical book summary videos with AI images, TTS voice, Ken Burns effect, and precise subtitle and audio alignment using customizable temp...

This skill appears to do what it says, but a few practical cautions before you run it: 1) Provide the API key only for the image service you expect (ARK / Volces) and verify the endpoint if you have doubts. 2) The script will attempt to pip-install edge-tts and requests at runtime — run it in a virtualenv or sandbox if you want to limit package installation. 3) Ensure ffmpeg is installed and available on PATH. 4) The script reads specific files (~/.qclaw/workspace/kdjojodsi.md and a couple of local filenames) to find a UUID-formatted key — do not store unrelated secrets in files with those names in your project directory. 5) Inspect the ARK service/provider (volces.com / 火山引擎) if you need to confirm trustworthiness. If any of these are concerns, run the code locally in an isolated environment or review/modify the script before use.

Type: OpenClaw Skill Name: book-video-maker Version: 2.0.0 The skill is a functional video generator, but it exhibits several high-risk behaviors in `scripts/generate.py`. It uses `subprocess` to automatically install Python dependencies via `pip` and to execute system binaries (`ffmpeg`, `ffprobe`) for media processing. Additionally, it searches for API keys in specific, non-standard file paths like `~/.qclaw/workspace/kdjojodsi.md` using regular expressions. While these actions are aligned with the stated purpose of generating videos with AI, the use of automated software installation and shell execution represents a significant attack surface without clear evidence of malicious intent.