Comfy-OrgComfy-OrgOfficialยท9 Agent Skills Included

comfy-cli

Generate images, video, and audio with ComfyUI from your terminal

Installs, runs, and manages ComfyUI so AI agents can generate images, videos, audio, and 3D content from simple commands. Eliminates manual workflow JSON editing, GPU setup headaches, and slow multi-shot video production through reusable recipes and cloud routing. Includes skills for debugging failed jobs, directing narrative videos, and previewing results directly in chat.
npx skills add Comfy-Org/comfy-cli --all -g -y

All Skills in This Repository (9)

Pure Emerald Level Indicators

Frequently Asked Questions

FAQPage Schema
How to install comfy-cli?โ–ผ

Run `npx skills add Comfy-Org/comfy-cli --all -g -y` in your terminal to install all skills in this suite globally.

How to generate AI images and video from the command line?โ–ผ

Use `comfy generate` for one-shot partner models like Flux or Kling, or run full ComfyUI workflows with `comfy run` locally or in the cloud.

Do I need a GPU to use comfy-cli?โ–ผ

No. Add `--where cloud` to route any workflow to Comfy Cloud, which runs jobs on hosted GPUs using your Comfy account credits.

Does comfy-cli work with Claude Code and Cursor?โ–ผ

Yes. The bundled skills teach Claude, Cursor, and other agents to drive the CLI natively, and every command outputs structured JSON for agents.

How to fix a failed ComfyUI workflow job?โ–ผ

Read the JSON error envelope's code and hint, then follow the comfy-debug skill, which maps every failure code to a specific fix.

Related Repositories in Design & Creative

View All in Design & Creativeโ†’