skills.rest
yakeworld
yakeworldCOMMUNITY
Explore all skills by this author →
GitHub

comfyui

Generate images and videos by executing ComfyUI API workflows against local or cloud servers.

6|1|Updated May 11, 2026
Search Tags:#workflow automation#api#image generation#video generation#text-to-image#comfyui#stable diffusion
One-click install
npx skills add https://github.com/yakeworld/Synthos --skill comfyui-yakeworld
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: comfyui
Source: https://github.com/yakeworld/Synthos/tree/main/skills/extended/external-automation/creative/tools/comfyui
Command: npx skills add https://github.com/yakeworld/Synthos --skill comfyui-yakeworld

SYSTEM DOCUMENTATION & REQUIREMENTS

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

What problem does it solve? Running ComfyUI image and video generation programmatically requires managing API-format workflows, verifying that models and custom nodes are installed, and handling local versus Comfy Cloud endpoint differences, which is error-prone when done manually. ## Core Features & Use Cases - Workflow Execution: Run API-format ComfyUI workflows (SD1.5, SDXL, Flux, AnimateDiff, Wan video) with prompt and parameter injection via run_workflow.py. - Dependency Checking: Verify required custom nodes, models, and embeddings against a running server with check_deps.py, and auto-install missing pieces with auto_fix_deps.py. - Parameter Discovery: Extract controllable parameters (prompt, seed, steps, cfg, dimensions) from any workflow using extract_schema.py. - Use Case: A user wants to generate a 1024x1024 SDXL image from a text prompt; the skill validates the workflow, checks the checkpoint is installed, executes it through the ComfyUI API, and returns image URLs with seed and dimension metadata. ## Quick Start Ask the agent to generate an image from a text prompt using the sdxl_txt2img workflow, optionally specifying seed, steps, and output directory.

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 the workflow JSON file and an --args JSON object containing parameters like prompt, seed, and steps. The script submits the workflow to the ComfyUI /prompt endpoint, polls for completion, and downloads outputs to the specified directory.

How do I check if a ComfyUI workflow's models and nodes are installed?▼

Run check_deps.py with the workflow file against your server. It queries /object_info and /models endpoints to detect missing custom nodes, checkpoints, LoRAs, and embeddings, and prints concrete install commands like comfy node install for each gap.

Does this work with Comfy Cloud or only local ComfyUI?▼

Both are supported. The scripts detect cloud hosts by domain, automatically add the /api prefix, map renamed endpoints like /history_v2 and /experiment/models, and authenticate with an API key from the COMFY_CLOUD_API_KEY environment variable.

What workflow format does ComfyUI API execution require?▼

Workflows must be in API format, where top-level keys are node IDs each containing a class_type field. Editor-format files with nodes and links arrays are rejected; re-export them from ComfyUI using Save (API Format).

Why does my ComfyUI workflow fail with missing model errors?▼

The workflow references checkpoints, LoRAs, or embeddings not present on the server, or uses local filenames while the cloud server uses -fp16 suffixed variants. Run check_deps.py to identify exact missing files and override names like ckpt_name through the args parameter.

Related Skills

Explore More Design & Creative Skills→
jakubbartnikjakubbartnik

comfyui

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

Community
Advanced
calesthiocalesthio
46.2k

ltx2

Generate video clips from text prompts or images using the LTX-2.3 22B DiT model.

Community
Advanced
luokai25luokai25

image-enhancer

Enhance image resolution, sharpness, and clarity through intelligent upscaling.

Community
Intermediate
HughYauHughYau
2.5k

using-model-endpoint

Call registered model endpoints over their native HTTP APIs from scoped inference kernels.

Community
Intermediate