Skip to main content
Glama

twitch-mcp

Twitch MCP Server

AI-powered tools for Twitch streamers, exposed via the Model Context Protocol (MCP). Connect your coding/chat assistants (Gemini CLI, Qwen Coder, Claude Code, etc.) to your Twitch chat for moderation, stream management, and engagement.

Quick Start (No Cloning Required)

Prerequisites

  • Node.js 14+ and Java 11+ available on your system PATH
  • A Twitch account with appropriate API credentials

Setup Configuration

Create a config file with your Twitch credentials (use bare access token, no "oauth:" prefix):

Windows: C:/Users/<you>/AppData/Roaming/twitch-mcp/config.json
macOS: ~/Library/Application Support/twitch-mcp/config.json
Linux: ~/.config/twitch-mcp/config.json

Example config.json:

{ "channel": "YOUR_TWITCH_USERNAME", "auth": "YOUR_TWITCH_ACCESS_TOKEN", "clientId": "YOUR_TWITCH_CLIENT_ID", "broadcasterId": "YOUR_BROADCASTER_ID", "showConnectionMessage": true }

🥇 Primary Recommendation: Gemini CLI

  • Excellent MCP support with reliable connections
  • Free tier with generous limits
  • Great for day-to-day Twitch moderation and chat management

🥈 Secondary Recommendation: Qwen Coder CLI

  • Strong coding-focused AI with good MCP integration
  • Particularly good for stream development and technical discussions
  • Free and open-source

⚡ For Power Users: Claude Code

  • Most advanced reasoning capabilities
  • Best for complex moderation decisions and nuanced chat analysis
  • Note: Limited credits - save for heavy lifting tasks

Configuration Examples

Gemini CLI Configuration

Add to your Gemini settings:

{ "mcpServers": { "twitch-mcp": { "type": "stdio", "command": "npx", "args": [ "-y", "twitch-mcp-server@latest", "--config", "C:/Users/<you>/AppData/Roaming/twitch-mcp/config.json" ] } } }
Qwen Coder Configuration

Add to your Qwen settings:

{ "mcpServers": { "twitch-mcp": { "type": "stdio", "command": "npx", "args": [ "-y", "twitch-mcp-server@latest", "--config", "C:/Users/<you>/AppData/Roaming/twitch-mcp/config.json" ] } } }

Features

  • Chat Management: Send and read messages, recent chat log, chat analysis
  • Moderation: Timeout/ban users (by username or descriptor keywords)
  • Stream Management: Update title/category, create clips
  • Interactive Tools: Create polls and predictions for viewer engagement

Installation Options

No installation required - your AI tool will automatically fetch the latest version:

npx twitch-mcp-server@latest --config /path/to/config.json

Option 2: Global Install

npm install -g twitch-mcp-server twitch-mcp-server --config /path/to/config.json

Option 3: Local Development

See README-developers.md for building from source.

Usage Notes

  • Use a bare access token in auth field (no oauth: prefix needed)
  • Only one AI client should connect at a time to avoid conflicts
  • Ensure Node.js and Java are installed on the same system as your AI CLI
  • The server will automatically handle IRC formatting and API authentication

Troubleshooting

Authentication Issues:

  • Confirm auth is a bare token (no oauth: prefix)
  • Verify clientId and scopes match your generated token
  • Ensure broadcasterId matches your channel ID

Connection Problems:

  • Check that both Node.js and Java are in your system PATH
  • Try running java -version and node -version to verify installation
  • For Windows users: Make sure you're not running from WSL if your AI CLI is on Windows

NPX Prompts in Headless Environments:

  • The configurations above use -y flag for non-interactive execution
  • If you encounter prompts, add --yes to the npx command

Support

  • For general usage questions, check the troubleshooting section above
  • For development and contributions, see README-developers.md
  • For Claude-specific integration help, see CLAUDE.md
Deploy Server
A
security – no known vulnerabilities
F
license - not found
A
quality - confirmed to work

hybrid server

The server is able to function both locally and remotely, depending on the configuration or use case.

AI-powered tools for Twitch streamers, exposed via the Model Context Protocol (MCP). Connect your coding/chat assistants (Qwen Code, LM Studio, Claude, etc.) to your Twitch chat for moderation, stream management, and engagement.

  1. Quick Start (No Cloning Required)
    1. Prerequisites
    2. Setup Configuration
    3. Recommended AI CLI Tools
    4. Configuration Examples
  2. Features
    1. Installation Options
      1. Option 1: NPX (Recommended)
      2. Option 2: Global Install
      3. Option 3: Local Development
    2. Usage Notes
      1. Troubleshooting
        1. Support

          Related MCP Servers

          • -
            security
            A
            license
            -
            quality
            A simple MCP server for interacting with OpenAI assistants. This server allows other tools (like Claude Desktop) to create and interact with OpenAI assistants through the Model Context Protocol.
            Last updated -
            36
            MIT License
            • Apple
          • A
            security
            F
            license
            A
            quality
            Enables interaction with the Twitch API, allowing users to retrieve comprehensive information about channels, streams, games, and more, with additional support for searching and accessing chat elements like emotes and badges.
            Last updated -
            14
            27
            1
            • Apple
            • Linux
          • A
            security
            A
            license
            A
            quality
            A Model Context Protocol (MCP) server that integrates Claude with the Terraform Cloud API, allowing Claude to manage your Terraform infrastructure through natural conversation.
            Last updated -
            62
            15
            MIT License
            • Linux
            • Apple
          • -
            security
            F
            license
            -
            quality
            An MCP server that allows AI assistants like Claude to execute terminal commands on the user's computer and return the output, functioning like a terminal through AI.
            Last updated -
            52
            • Apple

          View all related MCP servers

          MCP directory API

          We provide all the information about MCP servers via our MCP API.

          curl -X GET 'https://glama.ai/api/mcp/v1/servers/Eclipse-XV/twitch-mcp'

          If you have feedback or need assistance with the MCP directory API, please join our Discord server