YouTube Watcher

Fetch and read transcripts from YouTube videos. Use when you need to summarize a video, answer questions about its content, or extract information from it.

Security Scan

The skill does what it says — it uses yt-dlp to fetch YouTube subtitles and converts them to plain text; its requirements and behavior are consistent with its description.

Purpose & Capability

The skill's stated purpose is fetching and reading YouTube transcripts. It only requires the yt-dlp binary (and the provided script invokes yt-dlp) which is appropriate and expected.

Instruction Scope

SKILL.md instructs running the included Python script which runs yt-dlp in a temporary directory to fetch subtitles and prints cleaned text to stdout. The script only reads the temporary subtitle files it created and does not access other files, environment variables, or external endpoints beyond what yt-dlp uses (YouTube).

Install Mechanism

Install spec in the registry lists a brew formula for yt-dlp (reasonable). SKILL.md metadata also references a pip install option — a minor inconsistency but not malicious. Both brew and pip are common install methods for yt-dlp; there is no download-from-random-URL or arbitrary archive extraction.

Credentials

No environment variables, credentials, or config paths are requested. The script runs a subprocess and uses a temporary directory only — no secret access or unrelated credentials are required.

Persistence & Privilege

The skill is not always-enabled and does not request elevated or persistent privileges or modify other skills. It can be invoked by the agent (default), which is expected for a user-invocable skill.

Assessment

This skill appears coherent and limited to fetching YouTube subtitles using yt-dlp and converting them to plain text. Before installing: ensure you trust the source of yt-dlp (install via your package manager: brew or pip) since yt-dlp will make network requests to YouTube; the script prints transcript text to stdout (which your agent may send to the model), so avoid using it on videos containing sensitive/private information if you do not want that content sent to external services. Note the small metadata mismatch: registry lists brew install while SKILL.md mentions a pip option — that's benign but worth checking which install method you will use.

Fetch transcripts from YouTube videos to enable summarization, QA, and content extraction.

Usage

Get Transcript

Retrieve the text transcript of a video.

python3 {baseDir}/scripts/get_transcript.py "https://www.youtube.com/watch?v=VIDEO_ID"

Examples

Summarize a video:

1. Get the transcript:

   python3 {baseDir}/scripts/get_transcript.py "https://www.youtube.com/watch?v=dQw4w9WgXcQ"
   

2. Read the output and summarize it for the user.

Find specific information:

1. Get the transcript.
2. Search the text for keywords or answer the user's question based on the content.

Notes

• Requires yt-dlp to be installed and available in the PATH.
• Works with videos that have closed captions (CC) or auto-generated subtitles.
• If a video has no subtitles, the script will fail with an error message.