Setup & Installation
openclaw skills install @xtopher86/comfyui-runner
Or with the ClawHub CLI, for registry-managed skill folders outside a full OpenClaw workspace:
npx clawhub install comfyui-runner
What This Skill Does
Manages a local ComfyUI instance by starting, stopping, or checking its status. Connects via HTTP to the ComfyUI server using configurable host, port, and optional basic auth credentials.
Lets agents control ComfyUI lifecycle programmatically without a user manually opening a terminal or browser.
When to use it
- Starting ComfyUI before queuing image generation jobs
- Checking if ComfyUI is reachable from an agent workflow
- Stopping ComfyUI after a batch session to free GPU memory
- Verifying server status before triggering downstream image tasks
- Automating ComfyUI startup in a scheduled generation pipeline
Example Workflow
Here's how your AI assistant might use this skill in practice.
INPUT
User asks: Is my ComfyUI server running?
AGENT
- 1Reads COMFYUI_HOST and COMFYUI_PORT from environment or .env file
- 2Sends a curl request to the /health endpoint on the configured host and port
- 3Parses the HTTP response to determine reachability
- 4Returns server status to the user
OUTPUT
Server is reachable at 192.168.179.111:28188
Requirements
Accounts, API keys, or tools you or your AI assistant may need to set up while using this skill.
ComfyUI installed and running on an accessible host