Youtube Transcript Local
v1.0.1通過本地 yt-dlp 提取 YouTube 視頻多語言字幕，無需外部API，支持自動生成字幕和離線安全使用。
⭐ 0· 82·0 current·0 all-time
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
OpenClaw
Benign
high confidencePurpose & Capability
Name/description: extract YouTube transcripts locally via yt-dlp. The included Python code and SKILL.md call and manage yt-dlp, optionally ffmpeg/Whisper/Tesseract for audio/OCR. There are no unrelated credentials, binaries, or config paths requested.
Instruction Scope
SKILL.md and extract.py stay within the stated scope (run yt-dlp to fetch subtitles, optionally extract audio and run local STT/OCR). The skill does require network access to YouTube (documented in SKILL.md). It references other skills/tools (whisper, tesseract-ocr, browser screenshot) as fallbacks; those may expand runtime scope if used but are not part of this skill itself.
Install Mechanism
There is no explicit install spec in the registry, but extract.py will attempt to run 'pip install yt-dlp' at runtime if yt-dlp is not found. That means code will perform network installs from PyPI when executed — an operational risk (unexpected network package installation) though the package host (PyPI) is standard. No downloads from unknown custom URLs are present.
Credentials
The skill does not require environment variables or secrets. README suggests optional env vars for defaults (YT_DEFAULT_LANG, YT_OUTPUT_DIR, etc.); these are reasonable and not sensitive. The code writes local cache and output files in the working directory, which is expected behavior.
Persistence & Privilege
The skill does not request always:true and does not modify other skills or global agent config. It creates local cache/output files under the current working directory (./.cache, ./transcripts) but nothing else persistent or privileged is requested.
Assessment
This skill appears to do what it claims: locally fetch YouTube subtitles using yt-dlp and optionally use local STT/OCR fallbacks. Before installing/running: 1) Be aware extract.py will try to 'pip install yt-dlp' automatically if yt-dlp is not present — that triggers a network install from PyPI. If you prefer, pre-install yt-dlp and any optional tools (ffmpeg, Whisper, tesseract) in a virtualenv or container. 2) The tool needs network access to YouTube and will write files into the current working directory (creates ./transcripts and ./.cache) — run it in a directory you control. 3) The code is simple and uses subprocess.run without shell=True (safer), but you should still review or run it in an isolated environment if you don't trust the package source. 4) If you want stricter guarantees, clone the referenced GitHub repo yourself, inspect code locally, and install dependencies explicitly rather than relying on runtime pip installs.Like a lobster shell, security has layers — review code before you run it.
latestvk97fj9stt1p22yxdyc7wj9686s83eezs
License
MIT-0
Free to use, modify, and redistribute. No attribution required.