skills.rest
i-bebsi
i-bebsiCOMMUNITY
Explore all skills by this author →
GitHub

comfyui

Generate images, video, and audio by running ComfyUI workflows via REST and WebSocket APIs.

Updated Jun 17, 2026
Search Tags:#workflow automation#image generation#video generation#flux#comfyui#stable diffusion#comfy-cli
One-click install
npx skills add https://github.com/i-bebsi/hermes-agent --skill comfyui-i-bebsi
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: comfyui
Source: https://github.com/i-bebsi/hermes-agent/tree/main/hermes-config/skills/creative/comfyui
Command: npx skills add https://github.com/i-bebsi/hermes-agent --skill comfyui-i-bebsi

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires requests, websocket-client, comfy-cli, and includes scripts (resource) and references (resource) components.

What problem does it solve? Setting up ComfyUI and executing its workflows programmatically is error-prone: hardware requirements vary, workflows must be in API format, dependencies (custom nodes, models, embeddings) are often missing, and the official CLI lacks workflow execution support. This Skill automates the full lifecycle from hardware detection and installation to parameterized workflow execution and output retrieval. ## Core Features & Use Cases - Automated Setup & Health Checks: Detect GPU/VRAM, choose local install vs Comfy Cloud, install via comfy-cli, and verify with a one-line health check. - Workflow Execution with Parameter Injection: Extract controllable parameters from any API-format workflow, inject prompts/seeds/dimensions, submit jobs, monitor progress via WebSocket, and download outputs safely. - Dependency Management: Check workflows against a running server for missing custom nodes, models, and embeddings, then auto-install them with concrete fix commands. - Use Case: A user asks to generate 8 variations of an SDXL image with random seeds — the Skill validates the workflow, checks dependencies, runs a parallel batch, and returns the output file paths as JSON. ## Quick Start Ask the agent to generate an image with a prompt using the SDXL example workflow, and it will check the server, inject your prompt, run the workflow, and return the output files.

Frequently Asked Questions about comfyui

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

FAQPage Schema
How do I run a ComfyUI workflow from the command line?▼

Use run_workflow.py with an API-format workflow JSON and a JSON args object containing parameters like prompt, seed, and steps. The script submits the job to the server, monitors execution, and downloads outputs to your chosen directory.

How do I generate images with ComfyUI in the cloud without a GPU?▼

Sign up at comfy.org/cloud, generate an API key at platform.comfy.org, and set the COMFY_CLOUD_API_KEY environment variable. Then run workflows with --host https://cloud.comfy.org; note that running workflows via API requires a paid subscription.

What hardware do I need to run ComfyUI locally?▼

Local ComfyUI needs an NVIDIA GPU with at least 6 GB VRAM (8 GB for SDXL, 12 GB or more for Flux and video), an AMD GPU with ROCm on Linux, or an Apple Silicon Mac with at least 16 GB unified memory. Run hardware_check.py to get an automatic verdict.

Why does my ComfyUI workflow fail with class_type not found?▼

This error means a required custom node is not installed on the server. Run check_deps.py against the workflow to identify missing nodes and get the exact comfy node install command, or use auto_fix_deps.py to install them automatically.

Can I use editor-format workflow JSON with the ComfyUI API?▼

No, the API and these scripts require API-format JSON where each node has a class_type field. Editor-format files with top-level nodes and links arrays must be re-exported from the ComfyUI web UI using Workflow then Export (API).

How do I run img2img or inpainting with ComfyUI scripts?▼

Pass --input-image image=./photo.png to run_workflow.py, which uploads the image and injects its server-side filename into the workflow. For inpainting, pass a second --input-image mask_image=./mask.png alongside your prompt and denoise settings.

Related Skills

Explore More Design & Creative Skills→
nebiusnebius
17

cosmos3-setup

Configure the Cosmos3 framework by validating source repositories and staging model checkpoints.

Official
Advanced
phuhao00phuhao00
12

bilibili-operator

Generate Bilibili-specific copy and publish videos, images, and text posts.

Community
Intermediate
tharunramagiritharunramagiri

tarunai-connect-videocaptioner

Transcribe video speech and generate translated subtitles for media files.

Community
Advanced
NousResearchNousResearch
239.1k

comfyui

Generate images, video, and audio via ComfyUI diffusion workflows.

Official
Advanced