Most YouTube connectors stop at the Google Cloud console; this one works from the moment it is installed, which is the difference between using it now and adding it to a list of things to set up.
A narrow YouTube reader: give it a video and get back the title, description, channel, publish date and view count together with the transcript, using extraction libraries rather than the Data API, so there is nothing to sign up for.
- `get_yt_video_info` — the video's metadata and its transcript in one call
- Transcript handling with fallback logic that covers both manually created and auto-generated captions
- Logging throughout and error handling for the cases where a transcript simply is not available
Python 3.13+ and `pip install mcp-youtube-extract`, or pipx, then run `mcp_youtube_extract`. No configuration is needed — the metadata path works without credentials. `YOUTUBE_API_KEY` is optional and used only as an additional fallback for metadata when the extraction path fails; setting one means a Google Cloud project with the YouTube Data API v3 enabled and billing linked, against a free tier of 10,000 units per day.
One command — pip install mcp-youtube-extract