YouTube Transcript MCP Server

Extract transcripts, timestamps, and searchable text from any YouTube video via MCP. This server connects to YouTube's caption system and returns clean, structured transcript data for any compatible client.

What does YouTube Transcript MCP Server do?

This MCP server connects to YouTube's caption system, enabling you to:

• Extract full transcripts from any YouTube video with captions enabled
• Get timestamped transcripts for precise moment-by-moment reference
• Search within transcripts to find exactly when specific topics are discussed

Running on the Apify platform gives you automatic scaling, zero infrastructure management, and pay-per-use pricing.

Why use YouTube Transcript MCP Server?

• Research & summarization: Process hours of video content in seconds
• Content repurposing: Extract video content for blog posts, newsletters, or social media
• Meeting notes: Pull transcripts from recorded meetings and presentations
• Education: Search lecture videos for specific topics or concepts
• Competitive analysis: Analyze competitor YouTube content at scale
• Multilingual support: Fetch transcripts in any available language

How to use YouTube Transcript MCP Server

1. Get your Apify API token from Apify Console
2. Add this MCP server to your client's configuration
3. Start fetching YouTube transcripts

MCP client configuration

Add to your MCP client config (e.g. claude_desktop_config.json):

{
    "mcpServers": {
        "youtube-transcript": {
            "type": "url",
            "url": "https://YOUR_USERNAME--youtube-transcript-mcp.apify.actor/mcp",
            "headers": {
                "Authorization": "Bearer YOUR_APIFY_TOKEN"
            }
        }
    }
}

Tools

get_transcript

Returns the full plain-text transcript of a YouTube video. Ideal for summarization, analysis, and content extraction.

Input:

• videoUrl (required) — YouTube URL or video ID
• lang (optional) — Language code (e.g. "en", "es", "fr")

get_transcript_with_timestamps

Returns the transcript with [MM:SS] timestamps on each line. Great for referencing specific moments in a video.

Input:

• videoUrl (required) — YouTube URL or video ID
• lang (optional) — Language code

search_transcript

Searches the transcript for a keyword or phrase and returns only the matching segments with timestamps.

Input:

• videoUrl (required) — YouTube URL or video ID
• query (required) — Search term or phrase
• lang (optional) — Language code

Output example

{
    "content": [
        {
            "type": "text",
            "text": "[0:18] We're no strangers to love\n[0:22] You know the rules and so do I\n[0:27] A full commitment's what I'm thinking of"
        }
    ]
}

Pricing

Each transcript fetch costs $0.03. There are no monthly fees or minimum charges — you only pay when a tool is called.

Usage	Estimated cost
10 transcripts/day	~$9/month
100 transcripts/day	~$90/month
1,000 transcripts/day	~$900/month

Supported URL formats

• https://www.youtube.com/watch?v=VIDEO_ID
• https://youtu.be/VIDEO_ID
• https://youtube.com/embed/VIDEO_ID
• Raw video ID (11 characters)

Limitations

• Only works with videos that have captions/subtitles enabled
• Auto-generated captions may contain errors
• YouTube may rate-limit requests from a single IP under heavy use

FAQ

Is this legal? This server accesses publicly available YouTube caption data. It does not download or redistribute video content.

What if a video has no transcript? The tool returns a clear error message indicating no transcripts are available.

Can I get auto-generated captions? Yes — if a video only has auto-generated captions, those are returned by default.