mcpField
Y
youtube-mcp

youtube-mcp

Local
github.com/miller-joeSocial Media2Updated 20w ago

About youtube-mcp

youtube-mcp is an OAuth-authenticated MCP server for YouTube channel owners. It supports channel management, comment moderation, analytics, and an optional ComfyUI bridge that generates and sets thumbnails in one call.

Key features

  1. 01Edit video metadata and captions
  2. 02Reply to and moderate comments
  3. 03Manage playlists
  4. 04Query channel and Shorts analytics
  5. 05Generate and set thumbnails via ComfyUI

Use cases

  1. 01Update titles, descriptions, and tags
  2. 02Upload or remove caption tracks
  3. 03Review and respond to video comments
  4. 04Generate and publish a video thumbnail

MCP support

Exposes YouTube channel management and analytics operations to MCP clients using OAuth 2.0. When configured with ComfyUI, it also provides a tool to generate a thumbnail and set it on a video.

Configuration

Local server config is the same for every client. Copy it into your client’s MCP config file.

mcp.json · Universal config
{
"mcpServers": {
"miller-joe-youtube": {
"command": "npx",
"args": [
"-y",
"@miller-joe/youtube-mcp"
],
"env": {
"YOUTUBE_CLIENT_ID": "<YOUTUBE_CLIENT_ID>",
"YOUTUBE_CLIENT_SECRET": "<YOUTUBE_CLIENT_SECRET>",
"YOUTUBE_CLIENT_SECRET_FILE": "<YOUTUBE_CLIENT_SECRET_FILE>",
"YOUTUBE_TOKEN_FILE": "<YOUTUBE_TOKEN_FILE>",
"COMFYUI_URL": "<COMFYUI_URL>",
"COMFYUI_DEFAULT_CKPT": "<COMFYUI_DEFAULT_CKPT>"
}
}
}
}
Run commandnpx -y @miller-joe/youtube-mcp
Environment variables to set
YOUTUBE_CLIENT_IDGoogle OAuth 2.0 Client ID (from Google Cloud Console, Desktop app type). See README for setup.
YOUTUBE_CLIENT_SECRETSecretGoogle OAuth 2.0 Client Secret. Pair with YOUTUBE_CLIENT_ID.
YOUTUBE_CLIENT_SECRET_FILESecretAlternative to CLIENT_ID/CLIENT_SECRET: path to the JSON credentials file downloaded from Google Cloud.
YOUTUBE_TOKEN_FILESecretRefresh token storage path. Default: ~/.config/youtube-mcp/token.json
COMFYUI_URLComfyUI HTTP URL for the generate_and_set_thumbnail bridge tool. When unset, that tool is disabled.
COMFYUI_DEFAULT_CKPTDefault ComfyUI checkpoint filename for bridge tool (default: sd_xl_base_1.0.safetensors)

Local servers need a runtime on your machine (e.g. Node.js 18+). Config file locations differ by client; see your client’s docs.

Packages
npm@miller-joe/youtube-mcp0.1.2

Tools

No tool list yet

Local servers expose tools only once running on your machine, and the maintainer has not provided a tool list.

Details

From the project README

youtube-mcp

OAuth-authenticated YouTube MCP for channel owners. Edit video metadata, reply to and moderate comments, manage playlists, query channel analytics, and generate or set AI thumbnails via a ComfyUI bridge. Goes beyond the read-only Data API v3 wrappers that dominate this space.

youtube-mcp MCP server

GitHub Sponsors Ko-fi

The pitch

Most existing YouTube MCPs use an API key against Data API v3. Search videos, fetch public metadata, read-only. This one uses OAuth 2.0 (Authorization Code + PKCE) so it can actually write to your channel: update video titles, descriptions and tags, reply to comments, moderate spam, manage playlists. It also hits the separate YouTube Analytics API for channel stats, and generates a thumbnail via ComfyUI and pushes it to YouTube in a single MCP call.

Claude, use generate_and_set_thumbnail on video abc123:
  prompt: "cyberpunk hacker at keyboard, neon blue and pink, high contrast"

ComfyUI renders 1280×720, youtube-mcp fetches the bytes, and POSTs to thumbnails.set. Done.

Install

# npx, no install
npx @miller-joe/youtube-mcp --help

# Docker
docker run -p 9120:9120 \
  -e YOUTUBE_CLIENT_ID=... \
  -e YOUTUBE_CLIENT_SECRET=... \
  -e YOUTUBE_TOKEN_FILE=/token/token.json \
  -v $PWD/token:/token \
  ghcr.io/miller-joe/youtube-mcp:latest

Setup: Google Cloud one-time (~10 min)

  1. Google account plus YouTube channel. Use a personal account, not a workspace one you might lose.

  2. Google Cloud project at https://console.cloud.google.com. Call it whatever you want (e.g. youtube-mcp).

  3. Enable APIs:

    • YouTube Data API v3
    • YouTube Analytics API
  4. OAuth consent screen: External, App name, support email. In Scopes, add:

    • youtube.upload
    • youtube.force-ssl
    • yt-analytics.readonly
  5. Stay in Testing mode. Add yourself as a test user (required). As the project owner, your refresh token won't expire.

  6. Create OAuth Client ID: Application type = Desktop app. Download the JSON.

  7. Run the interactive auth flow:

    npx @miller-joe/youtube-mcp --auth --client-secret-file ./client_secret.json
    

    A browser opens, you log in to the Google account tied to your YouTube channel, and grant the requested scopes. On success, a refresh token is saved to ~/.config/youtube-mcp/token.json.

  8. Start the server:

    npx @miller-joe/youtube-mcp --client-secret-file ./client_secret.json
    

    Or provide the client credentials via env: YOUTUBE_CLIENT_SECRET_FILE, or YOUTUBE_CLIENT_ID + YOUTUBE_CLIENT_SECRET.

Connect an MCP client

claude mcp add --transport http youtube http://localhost:9120/mcp

Or point your MCP gateway at the Streamable HTTP endpoint.

Configuration

CLI flagEnv varDefaultNotes
--client-secret-fileYOUTUBE_CLIENT_SECRET_FILE(none)Path to Google OAuth JSON
--client-idYOUTUBE_CLIENT_ID(none)Alternative to the secret file
--client-secretYOUTUBE_CLIENT_SECRET(none)Alternative to the secret file
--token-fileYOUTUBE_TOKEN_FILE~/.config/youtube-mcp/token.jsonRefresh token storage
--hostMCP_HOST0.0.0.0Bind host (HTTP mode only)
--portMCP_PORT9120Bind port (HTTP mode only)
--stdioMCP_TRANSPORT=stdio(unset)Speak MCP over stdio instead of HTTP. Use when launched as a subprocess by a stdio-first MCP client (Claude Desktop, mcp-inspector).
--comfyui-urlCOMFYUI_URL(unset, bridge disabled)ComfyUI HTTP URL for bridge tools
(no flag)COMFYUI_DEFAULT_CKPTsd_xl_base_1.0.safetensorsDefault checkpoint for bridge tool
Transports

The server speaks streamable HTTP by default (great for Claude Code, MetaMCP, raw fetch). Pass --stdio (or set MCP_TRANSPORT=stdio) to switch into stdio mode, which is what stdio-first clients like Claude Desktop and the MCP Inspector expect:

// claude_desktop_config.json
{
  "mcpServers": {
    "youtube": {
      "command": "npx",
      "args": ["-y", "@miller-joe/youtube-mcp", "--stdio"],
      "env": {
        "YOUTUBE_CLIENT_SECRET_FILE": "/path/to/client_secret.json",
        "YOUTUBE_TOKEN_FILE": "/path/to/token.json"
      }
    }
  }
}

Stdio mode skips the OAuth-token preflight check — the server boots even without a stored token and surfaces auth errors at tool-call time. Run youtube-mcp --auth --client-secret-file <path> once in HTTP mode to seed the refresh token before pointing Claude Desktop at it.

Tools

Videos
  • list_my_videos: paginated list of the authenticated channel's uploads.
  • get_video: full detail for one video.
  • update_video_metadata: title, description, tags, category, privacy.
  • delete_video: permanently delete a video. Requires confirm_video_title to match the current title exactly, as a guard against deleting the wrong video.
Captions
  • list_captions: list caption tracks on a video (language, name, status, draft flag).
  • upload_caption: upload an SRT or WebVTT caption track to a video.
  • delete_caption: delete a caption track.
Shorts
  • list_my_shorts: find Shorts in recent uploads (filters by duration ≤60s).
  • get_shorts_analytics: YouTube Analytics query restricted to Shorts (creatorContentType==SHORTS).
Playlists
  • create_playlist: create a playlist (default private).
  • add_to_playlist: add a video to an existing playlist.
Comments
  • list_comments: top-level comment threads on a video.
  • reply_to_comment: reply to a top-level comment.
  • moderate_comment: hold, approve, or reject a comment.
Analytics
  • query_channel_analytics: date-ranged metrics with optional dimensions and filters.
Bridge (when COMFYUI_URL is configured)
  • generate_and_set_thumbnail: generate a thumbnail via ComfyUI and set it on a video in one call.

Quota notes

YouTube Data API free tier = 10,000 units/day. Key operation costs:

  • videos.list, commentThreads.list: 1 unit each.
  • videos.update, comments.insert, thumbnails.set: 50 units each.
  • videos.insert (upload): 1,600 units, so about 6 uploads per day on the free tier.

Most creator-ops workflows stay well under the free cap.

Architecture

┌────────────────┐     ┌──────────────────┐     ┌─────────────────┐
│  MCP client    │────▶│  youtube-mcp     │────▶│  YouTube APIs   │
│  (Claude etc.) │◀────│  (this server)   │◀────│  (Data/Analytics)│
└────────────────┘     └────────┬─────────┘     └─────────────────┘
                                │
                                │ (bridge tools only)
                                ▼
                       ┌──────────────────┐
                       │  ComfyUI         │
                       │  (txt2img)       │
                       └──────────────────┘

OAuth refresh tokens are cached locally and refreshed just-in-time before expiry. The bridge tool downloads image bytes from ComfyUI internally, so ComfyUI does not need to be publicly reachable.

Development

git clone https://github.com/miller-joe/youtube-mcp
cd youtube-mcp
npm install
npm run dev
npm run build
npm test

Requires Node 20+.

Roadmap

Shipped:

  • Videos: list, get, update metadata, delete with title-match confirm guard.
  • Captions: upload, list, delete.
  • Shorts: list_my_shorts (duration filter) and get_shorts_analytics (creatorContentType==SHORTS).
  • Playlists: create, add-to.
  • Comments: list, reply, moderate.
  • Analytics: channel analytics query.
  • ComfyUI thumbnail bridge: generate_and_set_thumbnail.

Planned:

  • Video upload (video_upload) with resumable-upload support.
  • Reporting API for bulk historical data exports.

License

MIT © Joe Miller

Support

If this saves you time, consider supporting development:

GitHub Sponsors Ko-fi

Comments

Markdown supported

New to MCP?

From what the protocol is to installing your first server, these answers cover most first questions.

An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.