🎩 You're Invited:Meet the Socket team at Black Hat in Las Vegas, August 3-6.RSVP
Sign In

youtube-organic-mcp

Package Overview
Dependencies
Maintainers
1
Versions
1
Alerts
File Explorer

Advanced tools

Socket logo

Install Socket

Detect and block malicious and high-risk dependencies

Install

youtube-organic-mcp

MCP server for YouTube organic analytics — channel stats, video performance, watch time, and audience insights via the YouTube Data & Analytics APIs. Built for Claude Code and other MCP-compatible AI tools.

latest
Source
npmnpm
Version
1.0.0
Version published
Weekly downloads
36
33.33%
Maintainers
1
Weekly downloads
 
Created
Source

youtube-organic-mcp

MCP server for YouTube organic analytics — channel stats, video performance, watch time, and audience engagement via the YouTube Data API v3 and YouTube Analytics API v2.

Built for Claude Code and any MCP-compatible AI tool. Gives your AI assistant direct, read-only access to your own channel's organic data — subscribers, views, watch time, per-video engagement, and day-by-day trends.

Part of The SEO Engine toolkit by Rex Jones — AI-powered SEO and social media tooling for agencies and businesses.

Why this exists

  • No good open-source YouTube organic MCP existed. Plenty of ad MCPs; organic channel analytics, nobody.
  • YouTube has the richest organic API of the major platforms. Watch time, average view duration, subscribers gained/lost, traffic sources — none of it gated behind a multi-week review.
  • Reuses your existing Google OAuth. If you already run the GA4 / GSC / Google Ads MCPs, the same OAuth client works here.

Tools

ToolDescription
list_accountsList configured channels and the default
get_channel_infoChannel profile + lifetime stats (subscribers, total views, video count, uploads playlist)
get_videosPaginated uploads with per-video views, likes, comments, duration
get_video_detailsDetails for specific video IDs (batch up to 50)
get_channel_analyticsChannel analytics over a date range: views, watch time, avg view duration/%, subs gained/lost, likes, comments, shares — optionally by day
get_video_analyticsThe same metrics for a single video over a date range

All tools accept an optional account parameter for multi-account setups.

Getting connected

Step 1: Enable the APIs

In Google Cloud Console (any project — you can reuse an existing one), enable YouTube Data API v3 and YouTube Analytics API.

Step 2: Get a refresh token

Create an OAuth client of type Desktop app and download its client_secret JSON. Then run the helper (needs pip install google-auth-oauthlib):

python3 scripts/get_youtube_token.py /path/to/client_secret.json my-channel

Authorize with the Google account that owns the channel (pick the Brand Account if prompted). It prints the three YOUTUBE_* values to drop into your config and writes them to .youtube_account.json (gitignored).

Step 3: Configure

Add to your Claude Code MCP settings:

{
  "mcpServers": {
    "youtube-organic": {
      "command": "node",
      "args": ["/path/to/youtube-organic-mcp/dist/index.js"],
      "env": {
        "YOUTUBE_CLIENT_ID": "your_client_id.apps.googleusercontent.com",
        "YOUTUBE_CLIENT_SECRET": "your_client_secret",
        "YOUTUBE_REFRESH_TOKEN": "1//your_refresh_token"
      }
    }
  }
}

Or set YOUTUBE_ACCOUNTS to a JSON array for multiple channels (see .env.example).

Architecture

src/
  index.ts          # MCP server entry point, tool registration
  accounts.ts       # Multi-account resolution (client_id/secret/refresh_token)
  client.ts         # OAuth token minting + Data/Analytics API HTTP client
  types.ts          # TypeScript interfaces for API responses
  tools/
    channel.ts      # get_channel_info + uploads-playlist resolver
    videos.ts       # get_videos, get_video_details
    analytics.ts    # get_channel_analytics, get_video_analytics
  • Zero external HTTP dependencies — native fetch (Node 18+)
  • Auto-refreshing auth — access tokens minted from the refresh token, cached per process
  • Multi-account support — monitor multiple channels from one server

License

MIT

Keywords

mcp

FAQs

Package last updated on 29 Jun 2026

Did you know?

Socket

Socket for GitHub automatically highlights issues in each pull request and monitors the health of all your open source dependencies. Discover the contents of your packages and block harmful activity before you install or update your dependencies.

Install

Related posts