luma

Luma MCP Server exposes Luma AI's video generation capabilities as an MCP service. Once running, you can invoke the available MCP tools to generate videos from prompts, create videos from a starting frame image, extend existing video outputs, or interpolate between two videos. The tools demonstrate typical MCP tool patterns, including a name and a set of arguments (such as prompt, image_url, and optional callback_url) to control behavior. The server handles validation, error mapping, and communicates with the Luma API to produce the requested video results. When integrating, supply prompts in English and consider using the callback_url to receive asynchronous results or progress updates.

Common workflows include:

• generate_video: produce a new video from a text prompt.
• generate_video_from_image: start from a specific image frame and continue generation using a prompt.
• extend_video: extend an existing video output using a continuation prompt and the source_generation_id.
• interpolate_videos: smoothly interpolate between two existing video generations by providing their IDs.

To use these tools, construct a request object with the tool name and an arguments block as shown in examples for each method, and send it to the MCP server endpoint configured by the running instance.

Prerequisites:

• Node.js and npm installed on your machine (recommended: LTS version).
• Access to install or run npm packages from the command line.

Step-by-step:

1. Ensure Node.js and npm are installed. Verify with:

   node -v npm -v

2. Install and run the Luma MCP Server via npx (as shown in the README):

# Install and start the MCP server via npx
npx -y @sunwood-ai-labs/luma-mcp-server

3. Configure environment variables:

# Set your Luma API key (adjust to your shell)
export LUMA_API_KEY=your_api_key_here

4. Verify the server is running and reachable by sending a test MCP request or by checking logs/output for startup confirmation.

5. If you need to run the server in a persistent environment, consider using a process manager or containerization as appropriate for your deployment target.

Notes and tips:

• Ensure the LUMA_API_KEY is kept secret and not committed to version control.
• Prompts should be written in English for best results.
• Video generation can be time-consuming; consider using the callback_url to receive results asynchronously.
• The MCP server relies on the Luma API; watch for API rate limits and error handling mappings in the server logs.
• If you encounter connectivity issues, verify network access to Luma API endpoints and that the API key is valid.
• The provided tools support optional loop behavior and callback URLs; use them to tailor the output persistence and notifications to your workflow.