io.github.runcomfy-com/runcomfy-mcp

Manage ComfyUI deployments and run async inference via RunComfy Serverless API.

Source

Official Registry

Updated

2026-09-07

// securityCaution

Remote endpoint· mcp.runcomfy.com
Runs locally
Official / curated
Open source
Community-trusted

Running security audit…

Credentials
Network egress
Code execution
Data access
Supply chain

// install

Copy the install command and let the AI configure it · recommended for beginners

Please install the "io.github.runcomfy-com/runcomfy-mcp" MCP server from askskill:
Run: claude mcp add --transport http 'io-github-runcomfy-com-runcomfy-mcp' 'https://mcp.runcomfy.com/mcp'

// use cases

Connect a ComfyUI workflow to remote serverless deployment

Developers or DevOps users can use this MCP tool to manage RunComfy remote deployment capabilities and connect local or existing ComfyUI workflows to a serverless API. This reduces manual API handling.

Handle inference requests asynchronously

When image generation or workflow execution is not suitable for synchronous waiting, this tool can submit async inference jobs and track status. It fits queued, polled, or background-processing scenarios.

// docs

No documentation provided

Check the source repo for usage and examples.

Go to source repo ↗

// FAQ

What does this tool do?

This is a remote MCP server for the RunComfy Serverless API, focused on ComfyUI deployment management and async inference. Based on the provided description, its core capabilities are deployments and async inference.

What prerequisites does it need?

From the available material, we can only confirm that it targets the RunComfy Serverless API. Whether it requires an API key, account, or specific runtime is not stated here; see the source repository.

How is it different from running ComfyUI locally?

Based on the description, it provides a remote MCP interface focused on serverless deployments and async inference rather than purely local execution. For more specific differences, see the source repository.