Skip to content

Tubask MCP Server ★ New (July 2 PM)

Overview

Tubask MCP (tubask.app/mcp · Amorizz/tubask-mcp) is a hosted YouTube MCP server designed for Claude, Cursor, and any MCP-compatible client. Search YouTube videos, extract full transcripts, retrieve metadata (titles, descriptions, view counts, publish dates), and analyze video content — all from a remote endpoint with zero infrastructure to manage. Content researchers, marketing teams, and operators who work with YouTube data can now query and analyze video content directly from their AI agents.

Key advantage: Hosted endpoint — no API keys to manage, no infrastructure, designed for Claude & Cursor but works with any MCP client.

Key Features

  • YouTube search: Search videos by keyword, channel, or topic — returns structured results with metadata
  • Transcript extraction: Extract full video transcripts with timestamps — support for auto-generated and manual captions
  • Video metadata: Title, description, view count, like count, publish date, duration, channel info
  • Channel analysis: Retrieve channel statistics, recent uploads, and content themes
  • Content summaries: AI-powered video content summarization for quick insights
  • Multi-language: Transcript extraction across major languages with auto-detection

Installation

# Remote endpoint — no local install needed
hermes mcp add tubask --url https://tubask.app/mcp

# Or self-hosted
git clone https://github.com/Amorizz/tubask-mcp.git
cd tubask-mcp
npm install
npm run build
hermes mcp add tubask --command "node" --args "dist/index.js" --workdir "$(pwd)"

Configuration

{
  "mcpServers": {
    "tubask": {
      "url": "https://tubask.app/mcp",
      "transport": "sse"
    }
  }
}

For self-hosted, add YouTube API key:

{
  "mcpServers": {
    "tubask": {
      "command": "node",
      "args": ["dist/index.js"],
      "workdir": "/path/to/tubask-mcp",
      "env": {
        "YOUTUBE_API_KEY": "your-youtube-data-api-key"
      }
    }
  }
}

Business Relevance

  • Content Research: Marketing teams can analyze competitor YouTube content — "extract transcripts from top 10 videos about 'AI customer service' and identify common themes"
  • Competitive Intelligence: Track competitor video publishing cadence, content themes, and audience engagement metrics over time
  • Video SEO: Analyze top-performing videos in your niche — extract titles, descriptions, and tags to inform your own video SEO strategy
  • Content Repurposing: Extract transcripts → summarize → convert to blog posts, social threads, newsletters
  • Market Research: "What are customers saying about [product] in YouTube reviews?" — sentiment analysis from video transcripts at scale

Integration with CorpusIQ

Tubask MCP enhances CorpusIQ's content and research capabilities, complementing existing YouTube and content tools:

  • Content Pipeline: Tubask (transcripts) → HumanTone (humanize) → Blog2Video (repurpose) — AI-driven content recycling
  • Competitive Monitoring: Combine with SEO tools (SE Ranking, OpenSEO) — track both video and web presence of competitors
  • Lead Intelligence: Tubask (competitor video analysis) → Google Maps Email Extractor (maps leads) → SyncGTM (enrichment) — multi-source lead intelligence
  • Market Research: Pair with Paper Search MCP for academic context — video content analysis + academic research = comprehensive market intelligence
  • Social Publishing: Extract insights → draft posts → publish via FeedSquad or Solnk — content insight to publication pipeline

Limitations

  • Hosted endpoint may have rate limits and usage quotas
  • Self-hosted requires YouTube Data API key (free quota: 10,000 units/day)
  • Transcript quality depends on YouTube's auto-captioning for non-manually-captioned videos
  • YouTube API quota costs apply at scale — self-hosted with own API key recommended for heavy usage

See Also