yutu
Verified Safeby eat-pray-ai
Overview
Automates YouTube workflows and manages YouTube resources as a command-line interface (CLI) tool and an MCP (Model Context Protocol) server for AI agents.
Installation
docker run --rm -it -u $(id -u):$(id -g) -v /absolute/path/to/credentials_dir:/app -p 8216:8216 -e YUTU_CREDENTIAL=/app/client_secret.json -e YUTU_CACHE_TOKEN=/app/youtube.token.json ghcr.io/eat-pray-ai/yutu:latest mcp --mode httpEnvironment Variables
- YUTU_CREDENTIAL
- YUTU_CACHE_TOKEN
- YUTU_LOG_LEVEL
- GEMINI_MODEL
- GEMINI_API_KEY
- YUTU_AGENT_INSTRUCTION
Security Notes
Relies on the user to securely handle OAuth credentials (client_secret.json and youtube.token.json). The tool's OAuth flow includes state verification for CSRF protection and uses a localhost redirect URI. Environment variables for credentials are correctly marked as secrets in the MCP server configuration. File operations for credentials use a controlled root directory.
Similar Servers
gcloud-mcp
Enables AI assistants to interact with the Google Cloud environment using the gcloud CLI for natural language commands, workflow automation, and simplified cloud management.
mcp-server-youtube-transcript
Retrieves transcripts from YouTube videos, including Shorts, with language fallback and optional ad filtering.
yt-dlp-mcp
This MCP server enables AI agents to interact with video platforms by providing tools for searching, extracting metadata, downloading content (video/audio), and generating transcripts/subtitles using the yt-dlp utility.
gofetch
A Model Context Protocol (MCP) server for efficiently retrieving, processing, and formatting web content from URLs.