Video Dub Clawhub
ReviewAudited by ClawScan on May 10, 2026.
Overview
The video-dubbing pipeline mostly matches its stated purpose, but it includes under-documented support for using browser cookies/session data for downloads, so credential handling should be reviewed before use.
Review credential handling before installing. Provide API keys only in the current session when possible, avoid enabling browser-cookie import unless you fully understand it, prefer a scoped cookies.txt file for the intended video site, and run the pipeline in an isolated Python environment.
Findings (4)
Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.
If enabled, the pipeline may use logged-in browser sessions to access video platforms, which can expose account cookies beyond a simple video URL download.
This enables yt-dlp to use browser cookie/session data when the environment variable is set. Browser cookie/profile access is high-impact credential handling, and the provided SKILL.md requirements document a cookies.txt file but do not clearly bound or explain browser-profile cookie use.
cookies_from_browser = os.getenv("YTDLP_COOKIES_FROM_BROWSER", "").strip() ... cookies_option["cookiesfrombrowser"] = (cookies_from_browser,)Prefer an explicitly exported cookies.txt file limited to the intended site. If browser-cookie import remains supported, document the exact env var, supported browsers/profiles, what cookies are used, and require explicit user approval before use.
Private or sensitive video content may be sent to translation or TTS providers during normal operation.
The translation and TTS stages necessarily involve external provider services. This is purpose-aligned and disclosed, but it means transcript text and generated speech requests may leave the local machine.
`DEEPSEEK_API_KEY` ... `DeepSeek API key for translation` ... `Edge TTS (default)` ... `requires no API key but an internet connection`
Use this only with content you are comfortable sending to the selected providers, and review each provider’s data-use terms before processing sensitive videos.
Running the pipeline will execute local commands and process downloaded media files, which may consume disk, CPU, GPU, and network resources.
The skill runs local media and download tools, which is expected for a video-processing pipeline. The artifacts describe this behavior rather than hiding it.
**`subprocess`**: Used to call ffmpeg, ffprobe, and yt-dlp for video processing.
Run it in a dedicated working directory or virtual environment, and only process URLs and media files you trust.
Future installs may receive newer package versions with changed behavior or vulnerabilities.
Dependencies are installed from package indexes using lower-bound version ranges rather than pinned hashes or an included lockfile. This is common but leaves more supply-chain variability.
yt-dlp>=2025.1.26 openai-whisper>=20240930 torch>=2.2.0 openai>=1.30.0 requests>=2.31.0
Install in a virtual environment and consider pinning exact package versions or using a lockfile before production use.