MCP Server with RunwayML Integration

This MCP (Model Context Protocol) server provides calculator tools and comprehensive RunwayML API integration for AI-powered video and image generation.

Features

Calculator Tools

• add: Simple addition of two numbers
• calculate: Multi-operation calculator (add, subtract, multiply, divide)

RunwayML API Tools

• runway_text_to_video: Generate videos from text prompts
• runway_image_to_video: Generate videos from images with text prompts
• runway_text_to_image: Generate images from text prompts with optional reference images
• runway_get_task: Check the status and retrieve results of generation tasks
• runway_cancel_task: Cancel running generation tasks
• runway_get_organization: Get organization information and credits

🖥️ Claude Desktop App Installation

Quick Setup (Recommended)

Run the automated setup script:

./setup-claude.sh

This script will:

• Install dependencies
• Create Claude configuration automatically
• Test the server
• Provide next steps

Manual Setup

For detailed manual installation instructions, see: CLAUDE_INSTALLATION.md

What You Get

Once installed in Claude Desktop App, you can:

• Generate RunwayML videos and images directly in Claude conversations
• Use calculator tools for quick math
• Check RunwayML account status and credits
• Monitor generation task progress

Example Claude conversation:

You: "Create a 10-second video of a cat playing in a garden"

Claude: Uses runway_text_to_video tool to generate the video

☁️ Cloudflare Workers Deployment

This server can also be deployed as a Cloudflare Worker for web-based access.

Setup

Prerequisites

1. Node.js and npm installed
2. A RunwayML account and API key
3. Cloudflare Workers account (for deployment)

Installation

npm install

Getting a RunwayML API Key

1. Sign up at RunwayML Developer Portal
2. Create an organization
3. Generate an API key in the API Keys tab
4. Add credits to your organization (minimum $10)

🔑 API Key Management

You have several options for providing your RunwayML API key:

Option 1: Pass as Parameter (Recommended for Testing)

Include the API key directly in each tool call:

{
  "prompt": "A serene lake at sunset",
  "api_key": "your_runway_api_key_here"
}

Option 2: Environment Variables (Local Development)

Set the API key as an environment variable:

For local development:

# In your terminal
export RUNWAYML_API_KEY="your_runway_api_key_here"
npm run dev

Or create a .env file:

# Create .env file in project root
echo "RUNWAYML_API_KEY=your_runway_api_key_here" > .env

Option 3: Cloudflare Workers Environment Variables (Production)

For production deployment, set environment variables in Cloudflare Workers:

Using Wrangler CLI:

# Set the environment variable for your worker
wrangler secret put RUNWAYML_API_KEY
# Enter your API key when prompted

Or via Cloudflare Dashboard:

1. Go to your Cloudflare Workers dashboard
2. Select your worker
3. Go to Settings → Environment Variables
4. Add RUNWAYML_API_KEY with your API key value

Option 4: wrangler.toml Configuration (Not Recommended for Security)

You can add environment variables to wrangler.toml, but this is not recommended for sensitive data:

[env.production.vars]
RUNWAYML_API_KEY = "your_key_here"  # Don't do this for real keys!

🚀 Development

Development

npm run dev

Deployment

npm run deploy

RunwayML API Usage

Text-to-Video Generation

Generate videos from text descriptions using RunwayML's latest models.

Parameters:

• prompt: Text description of the video you want to generate
• model: Choose between "gen4_turbo" (latest) or "gen3a_turbo"
• duration: Video length in seconds (5-10)
• ratio: Aspect ratio ("1280:720", "1920:1080", "720:1280", "1080:1920")
• api_key: Your RunwayML API key

Example:

{
  "prompt": "A serene lake at sunset with mountains in the background",
  "model": "gen4_turbo",
  "duration": 5,
  "ratio": "1920:1080",
  "api_key": "your_api_key_here"
}

Image-to-Video Generation

Animate existing images with text prompts.

Parameters:

• prompt_image: URL or base64 data URI of the input image
• prompt_text: Text description of how the image should be animated
• model: Choose between "gen4_turbo" or "gen3a_turbo"
• duration: Video length in seconds (5-10)
• ratio: Aspect ratio
• api_key: Your RunwayML API key

Example:

{
  "prompt_image": "https://example.com/image.jpg",
  "prompt_text": "The leaves gently swaying in the wind",
  "model": "gen4_turbo",
  "duration": 5,
  "ratio": "1920:1080",
  "api_key": "your_api_key_here"
}