runway-api
MCP server from runwayml/runway-api-mcp-server
claude mcp add --transport stdio runwayml-runway-api-mcp-server node <ABSOLUTE_PATH_TO_YOUR_CLONED_REPO_FROM_STEP_1>/build/index.js \ --env MCP_TOOL_TIMEOUT="1000000" \ --env RUNWAYML_API_SECRET="<YOUR_RUNWAY_API_KEY_HERE>"
How to use
This MCP server acts as a bridge to the Runway API, exposing a set of tools to generate and manipulate media via the Runway API. It enables creating images and videos from prompts, upscaling videos, editing videos with optional reference images, and querying task or organization information through dedicated endpoints. You can invoke tools like runway_generateImage, runway_generateVideo, runway_upscaleVideo, runway_editVideo, runway_getTask, runway_cancelTask, and runway_getOrg to perform common Runway API operations from within your MCP-enabled environment.
To use it, configure the server in your MCP client (e.g., Claude Desktop) to run the Node.js server that serves the tool endpoints. You’ll typically pass your Runway API key as RUNWAYML_API_SECRET and can adjust the MCP tool timeout with MCP_TOOL_TIMEOUT. Once configured, Claude (or your MCP client) will present the Runway tools under the runway-api-mcp-server namespace, allowing you to generate media by supplying the required prompts, references, and parameters as listed in the tool descriptions.
How to install
Prerequisites:
- Node.js installed
- Access to a Runway API key
- MCP workflow tooling (e.g., Claude Desktop) if you plan to integrate with Claude
Install steps:
- Clone the repository locally.
- Install dependencies: npm install
- Build the project to generate the build artifacts: npm run build You should see a build/index.js file after this step.
- Run the MCP server using Node.js (example): node build/index.js
Configuration for Claude Desktop (example):
- In Claude’s MCP config, add: { "mcpServers": { "runway-api-mcp-server": { "command": "node", "args": [ "<ABSOLUTE_PATH_TO_YOUR_CLONED_REPO_FROM_STEP_1>/build/index.js" ], "env": { "RUNWAYML_API_SECRET": "<YOUR_RUNWAY_API_KEY_HERE>", "MCP_TOOL_TIMEOUT": "1000000" } } } }
After saving, restart Claude Desktop to load the runway-api-mcp-server tools.
Additional notes
Notes and tips:
- Your Runway API key should be kept secret and provided via RUNWAYML_API_SECRET.
- The MCP_TOOL_TIMEOUT environment variable can be adjusted if you need longer-running tasks.
- Images generated via Runway API links are only available for 24 hours; download assets if you need to preserve them.
- Ensure your Runway API account has the necessary permissions for the actions you’re performing (generate, edit, upscale, etc.).
- If you encounter authentication or network issues, verify that RUNWAYML_API_SECRET is correctly set and that your server can reach Runway’s API endpoints.
Related MCP Servers
iterm
A Model Context Protocol server that executes commands in the current iTerm session - useful for REPL and CLI assistance
mcp
Octopus Deploy Official MCP Server
furi
CLI & API for MCP management
editor
MCP Server for Phaser Editor
DoorDash
MCP server from JordanDalton/DoorDash-MCP-Server
mcp
MCP сервер для автоматического создания и развертывания приложений в Timeweb Cloud