Allows extracting, searching, and analyzing YouTube video transcripts, supporting full transcript retrieval, keyword searching with context, time-chunked summaries, and batch processing for multiple videos.
Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type @ followed by the MCP server name and your instructions, e.g., "@YouTube Transcript MCP Server get the transcript for https://www.youtube.com/watch?v=dQw4w9WgXcQ"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
An MCP (Model Context Protocol) server that extracts, searches, and analyzes YouTube video transcripts. Works with Claude Desktop, Cursor, and any MCP-compatible client.
Get Transcript - Extract full transcript from any YouTube video
Search Transcript - Find specific keywords with surrounding context
Transcript Summary - Get time-chunked transcript for easier analysis
Batch Processing - Process up to 10 videos at once
Related MCP server: YouTube Insights MCP Server
pip install yt-transcript-mcpgit clone https://github.com/alex2zimmermann-ux/yt-transcript-mcp
cd yt-transcript-mcp
pip install .Add to your claude_desktop_config.json:
{
"mcpServers": {
"youtube-transcript": {
"command": "yt-transcript-mcp",
"env": {
"YT_MCP_MODE": "standalone"
}
}
}
}Variable | Default | Description |
|
|
|
|
| Backend service URL |
| - | API key for backend |
|
| Max cache entries |
|
| Cache TTL in seconds |
|
| Rate limit |
|
|
|
Uses youtube-transcript-api directly. Lightweight, no external dependencies. Best for marketplace deployment.
Connects to a running transcript service (FastAPI) that supports cookies, yt-dlp, and Whisper fallback. Best for premium/self-hosted usage.
get_transcriptGet the transcript of a YouTube video.
Parameters:
url (required) - YouTube URL or video ID
language (optional, default: "en") - Language code
format (optional, default: "text") - "text", "segments", or "both"
search_transcriptSearch for keywords in a video transcript.
Parameters:
url (required) - YouTube URL or video ID
query (required) - Search term
language (optional, default: "en")
context_segments (optional, default: 1) - Surrounding segments to include
get_transcript_summaryGet transcript in time chunks for analysis.
Parameters:
url (required) - YouTube URL or video ID
language (optional, default: "en")
chunk_minutes (optional, default: 5)
batch_transcriptsProcess multiple videos at once.
Parameters:
urls (required) - List of YouTube URLs or IDs (max 10)
language (optional, default: "en")
# Standalone
docker compose --profile standalone up
# Backend
YT_MCP_BACKEND_API_KEY=your-key docker compose --profile backend uppython -m venv .venv
source .venv/bin/activate
pip install -e ".[dev]"
pytest tests/ -v --covMIT
Get transcripts for multiple YouTube videos in a single request (max 10 videos).
Inputs: urls (required), language. Returns: result
Get the full transcript of a YouTube video in the specified language and format.
Inputs: url (required), language, format. Returns: result
Get a transcript organized into time-based chunks for easier analysis of long videos.
Inputs: url (required), language, chunk_minutes. Returns: result
Search for keywords or phrases in a YouTube video transcript and return matching segments with timestamps.
Inputs: url (required), query (required), language, context_segments. Returns: result
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
Scored across 4 tools
Each tool has a clearly distinct purpose: batch_transcripts handles multiple videos, get_transcript retrieves a full transcript, get_transcript_summary provides chunked analysis, and search_transcript finds specific segments. There is no overlap in functionality, making tool selection straightforward for an agent.
All tool names follow a consistent verb_noun pattern (e.g., get_transcript, search_transcript) with clear, descriptive verbs. The naming is uniform across all four tools, enhancing readability and predictability.
With 4 tools, the server is well-scoped for its purpose of YouTube transcript retrieval and analysis. Each tool serves a unique and essential function, avoiding bloat while covering core operations like retrieval, summarization, and search.
The tool set provides complete coverage for the domain: it supports single and batch transcript retrieval, summarization for long videos, and search capabilities. There are no obvious gaps, as all key workflows for transcript analysis are addressed without dead ends.
YouTube transcripts, search, channel browsing, and playlists for AI agents via MCP.
Real-time YouTube transcripts, comments, channels & video search MCP. Local npx or hosted OAuth.
An MCP server that gives any LLM or agent clean YouTube transcripts on demand: a single video, a whole channel, or a playlist, plus AI cleanup of auto-generated captions. API-key auth, credit-based, same backend as the public v1 API. Get a free API key with 25 free credits at youtubetranscriptdownload.com/account.
Transcripts, translations, summaries, subtitles, and search for public YouTube videos.