How do I use RunComfy MCP Server?

1. Click on "Install Server". 2. Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state. 3. In the chat, type @ followed by the MCP server name and your instructions, e.g., "@RunComfy MCP Server generate a video of a serene mountain landscape at sunset with a 16:9 aspect ratio" That's it! The server will respond to your query, and you can continue using it as needed. Here is a step-by-step guide with screenshots.

RunComfy MCP Server

MCP server to generate AI videos and images using the RunComfy APIs.

Setup

Install dependencies:

cd mcp-servers/runcomfy bun install

Get your API key from: https://www.runcomfy.com/profile
Add the configuration to your ~/.windsurf/mcp_config.json:

{ "mcpServers": { "runcomfy": { "command": "bun", "args": [".../mcp-runcomfy/index.js"], "env": { "RUNCOMFY_API_KEY": "YOUR_API_KEY_HERE" } } } }

Restart Windsurf

Available tools

runcomfy_generate_video

Generate an AI video. Parameters:

prompt (required): Video description
model: wan-2.1, wan-2.1-720p, animatediff, svd, kling, minimax
image_url: Image URL for image-to-video models
duration: Duration in seconds
aspect_ratio: 16:9, 9:16, 1:1
seed: Seed for reproducibility

runcomfy_generate_image

Generate an AI image (text-to-image). Parameters:

prompt (required): Image description
model: flux-2-pro
aspect_ratio: 16:9, 9:16, 1:1
seed: Seed for reproducibility
inputs: Advanced model-specific inputs (object)

runcomfy_edit_image

Edit an image (image-to-image). Parameters:

prompt (required): Edit instruction
model: flux-2-dev-edit, flux-kontext-pro-edit, qwen-edit-next-scene
image_url: Single image URL (some models)
image_urls: Multiple image URLs (some models)
aspect_ratio: 16:9, 9:16, 1:1
seed: Seed for reproducibility
inputs: Advanced model-specific inputs (object)

runcomfy_check_status

Check the status of a request.

runcomfy_get_result

Get the result (video URL) of a completed request.

runcomfy_cancel

Cancel a queued request.

runcomfy_list_models

List available models plus curated alias maps for video and image.

runcomfy_download_media

Download a generated media to local disk. Parameters:

url: Direct HTTPS URL to the media file
request_id: Resolve the media URL from a completed request (via /v1/requests/{request_id}/result)
kind: Preferred output kind when resolving from request_id (image or video)
index: Optional index if multiple outputs exist (0-based)
output_dir: Directory to save into (defaults to a temp folder)
filename: Optional filename override
overwrite: Overwrite if file exists (default false)
return_mode: How to return the downloaded media (default path)
- path: Return only JSON with saved_path
- resource_link: Also return a resource_link pointing to file://...
- embedded: Also return an embedded resource with base64 blob

Usage example

// Generate video runcomfy_generate_video({ prompt: "A calm person breathing slowly, teal glow, dark background", model: "wan-2.1", aspect_ratio: "1:1" }) // Check status runcomfy_check_status({ request_id: "abc123" }) // Get result runcomfy_get_result({ request_id: "abc123" }) // Generate image runcomfy_generate_image({ prompt: "A minimal flat illustration of a calm person breathing, teal and navy", model: "flux-2-pro", aspect_ratio: "1:1" }) // Edit image runcomfy_edit_image({ prompt: "Change the background to a dark gradient and add a subtle teal glow", model: "flux-2-dev-edit", image_urls: ["https://example.com/input.png"] }) // Download result media by URL runcomfy_download_media({ url: "https://example.com/output.png", output_dir: "/tmp", filename: "output.png", overwrite: true }) // Download result media by request_id runcomfy_download_media({ request_id: "abc123", kind: "image", index: 0 }) // Download and return a resource_link runcomfy_download_media({ request_id: "abc123", kind: "image", return_mode: "resource_link" }) // Download and return an embedded base64 resource runcomfy_download_media({ request_id: "abc123", kind: "image", return_mode: "embedded" })