search_youtube_videosLABSCO SUMMARY
Turns video into text you can search, quote and summarize.
Search and transcript together are what make a video answerable: you find the handful of videos that cover a topic, then read what was actually said in them — with timestamps, so a claim can be pointed back at the moment it was made.
What it is
A YouTube server with two jobs: search the platform for videos on a query, and pull the transcript of a video from its URL or ID.
What you get
- A video search with a result count you set
- Titles, channels, descriptions, URLs and metadata for the results
- A full transcript with timestamps for a video you name by URL or ID
- A built-in prompt template for analysing several videos on a topic at once
Requirements
A YouTube Data API v3 key from a Google Cloud project, set as an environment variable. Python, run from its own virtual environment.
Setup effort
Follow the published steps — python3 -m venv .venv
USE CASES
01
Getting the substance of a long video in a couple of minutes
Pull the transcript by URL and read or summarize it instead of scrubbing through the playback.
02
Surveying what has been said about a topic
Search for the videos, then read across their transcripts rather than watching each one.
03
Finding the exact moment someone said something
The transcript comes back with timestamps, so a quote can be linked to the second it appears.
04
Researching a channel before you commit time to it
Search with a tight result count and read titles, descriptions and metadata first.
3 tools
Listed in project docs. Your client may show a different list.
get_youtube_transcriptanalyze_youtube_content_promptBEFORE YOU INSTALL
What to know first
Tools you get
3 tools named in the project's documentation.
Requires
- Python 3 with venv and pip
- uv installed — the Claude Desktop entry starts the server with it
- YOUTUBE_API_KEY in your environment
Where it runs
The documented client runs the server on your machine.
INSTALLATION
Connect your client
Server setup
TERMINAL
git clone https://github.com/bendelpino/youtube-mcp
cd youtube-mcp
python3 -m venv .venv
source .venv/bin/activate
pip install -r requirements.txtCreate a Google Cloud project, enable YouTube Data API v3 and create an API Key credential
TERMINAL
cp .env.example .envPut YOUTUBE_API_KEY=<YOUR_YOUTUBE_API_KEY> in .env
Add the mcpServers block under MCP Clients to your Claude Desktop MCP settings
MCP Clients
Claude Desktop
CLAUDE DESKTOP · JSON
{
"mcpServers": {
"youtube": {
"command": "<YOUR_PATH_TO_UV>",
"args": [
"run",
"--directory",
"<YOUR_PATH_TO_PROJECT>",
"youtube_server.py"
],
"env": {
"YOUTUBE_API_KEY": "<YOUR_YOUTUBE_API_KEY>"
}
}
}
}<YOUR_PATH_TO_UV> is the full path of your uv executable (which uv prints it), and <YOUR_PATH_TO_PROJECT> is the full path of your youtube-mcp clone (the folder with youtube_server.py and the .venv from the steps).
