Overview
youtube-transcript-mcp is a tool that converts YouTube videos into Markdown transcripts for use with large language models. It fetches subtitles using yt-dlp and generates a structured Markdown file that includes metadata, an index of chapters, and the transcript text with timestamps. The tool is designed to reduce token usage by saving the transcript to disk and returning only a summary to the model. The transcription itself happens locally and does not incur model costs.
The server uses yt-dlp to fetch video metadata and caption tracks, then parses the subtitles into a structured Markdown document. It organizes the transcript into sections based on the video's chapters or fixed time blocks, and includes timestamps and links to YouTube at specific times. The output file includes YAML front matter with metadata such as title, channel, duration, and language. The tool provides a transcribe_video function that accepts parameters like URL, language, and output folder, and a list_languages function to check available caption tracks.
The tool is installed as an MCP server and can be connected to clients like Claude Code or Claude Desktop. It has limitations: it only works with videos that have captions, automatic captions may be inaccurate, and live streams must be finished before transcription. It also handles rate limits by backing off and suggesting browser cookies.
Pricing
Free
Detailed plans have not been confirmed in our catalog. Check the official website for current limits and billing terms.
Visit WebsitePrices and limits may change. Confirm the currency, billing period, seat minimum and usage caps on the official website.
Use Cases
- Researchers use it to get transcripts of talks for study and research.
- Developers use it to integrate YouTube transcripts into LLM workflows.
- Teams use it to create searchable notes from video content.
- Users use it to ask questions about specific parts of a video without loading the whole transcript.
- Users can ask the model to answer questions about specific topics in a video.
Features
Convert captioned videos into Markdown
Include timestamps in transcript text
Organize transcripts by video chapters
Add video metadata to transcripts
Link transcript passages to playback times
List available caption languages
Save transcripts for later analysis
Reviews
No reviews yet. Be the first to share your experience!