cli-anything-comfyui

Manage ComfyUI workflow queues and download generated images via REST API.

Updated May 7, 2026
One-click install
npx skills add https://github.com/dawsonblock/JUDGE_ATLAS --skill cli-anything-comfyui-dawsonblock
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: cli-anything-comfyui
Source: https://github.com/dawsonblock/JUDGE_ATLAS/tree/main/external/CLI-Anything-main/comfyui/agent-harness/cli_anything/comfyui/skills
Command: npx skills add https://github.com/dawsonblock/JUDGE_ATLAS --skill cli-anything-comfyui-dawsonblock

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

It solves the problem of manually operating ComfyUI for AI image generation by letting you queue workflows, monitor progress, and download results without using the GUI.

Core Features & Use Cases

  • Workflow orchestration: list saved workflows, load workflows from JSON, and validate workflow JSON against the ComfyUI node graph.
  • Queue control for generation runs: enqueue prompts, check running/pending status, clear the queue, view execution history, and interrupt the current generation.
  • Model and output management: discover checkpoints/LoRAs/VAEs/ControlNet models, list node types, list generated images, and download individual images or entire prompt outputs—optionally in machine-readable JSON for agents.

Quick Start

Ask your AI agent to queue a workflow for execution and return the queue status afterward using machine-readable output.

Frequently Asked Questions about cli-anything-comfyui

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I manage ComfyUI workflow queues from the command line?

You can manage ComfyUI workflow queues from the command line by using this Skill to enqueue prompts, check running and pending status, clear the queue, and interrupt current generations via the ComfyUI REST API without needing the GUI.

Can I download generated images from ComfyUI without using the web interface?

Yes, you can download generated images from ComfyUI without the web interface by executing image retrieval commands through this Skill to list generated images and download individual files or entire prompt outputs directly to your local machine.

Does this ComfyUI REST API tool support machine-readable JSON output for agents?

Yes, this ComfyUI REST API tool supports machine-readable JSON output for agents by appending the --json flag to commands, enabling AI-driven automation for workflow orchestration and queue monitoring.

How do I validate workflow JSON against the ComfyUI node graph?

To validate workflow JSON against the ComfyUI node graph, this Skill loads your provided workflow file and checks its structure and parameters to ensure compatibility before enqueuing it for execution on the server.

Do I need a running ComfyUI server to queue workflows via the command line?

Yes, you need a reachable ComfyUI server running locally at http://localhost:8188 to queue workflows, monitor execution, and retrieve generated images using this command line interface tool.

What is the best way to discover available checkpoints and LoRAs in ComfyUI?

The best way to discover available checkpoints, LoRAs, VAEs, and ControlNet models in ComfyUI is to use this Skill's model discovery commands, which query the server and list all currently installed and available models.