skills.rest
mooch10
mooch10COMMUNITY
Explore all skills by this author →
GitHub

comfyui

Generates images, video, and music through ComfyUI workflows in OpenMontage.

1|Updated Aug 28, 2026
Search Tags:#workflow#image generation#video generation#lora#comfyui#ace-step#openmontage
One-click install
npx skills add https://github.com/mooch10/mochiptos --skill comfyui-mooch10
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: comfyui
Source: https://github.com/mooch10/mochiptos/tree/main/frontend/.agents/comfyui
Command: npx skills add https://github.com/mooch10/mochiptos --skill comfyui-mooch10

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Running ComfyUI generation jobs from OpenMontage requires correct server configuration, API-format workflow exports, output node selection, and model placement, and mistakes in any of these cause failed or unreproducible generations. ## Core Features & Use Cases - Workflow Selection and Execution: Choose between bundled workflows and custom workflow_json/workflow_path inputs for comfyui_image, comfyui_video, and comfyui_music, including low-VRAM options like Wan 2.1 1.3B, LTXV FP8, and GGUF quantized workflows. - Output Node and Model Management: Identify the correct terminal saver node (SaveImage, SaveVideo, VHS_VideoCombine, SaveAudioMP3), place models in the proper ComfyUI directories, and handle missing models through the structured missing_models payload. - Failure Recovery and Provenance: Resume timed-out jobs with resume_prompt_id, record workflow hashes and model stacks for reproducibility, and configure per-capability server URLs. - Use Case: A user on a 12GB GPU wants to generate a short video from a community Wan 2.2 GGUF workflow; this Skill guides exporting it in API format, selecting the output_node, placing the required models, and submitting the job. ## Quick Start Use the comfyui skill to generate an image with the bundled workflow on my local ComfyUI server at localhost:8188.

Frequently Asked Questions about comfyui

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

FAQPage Schema
How do I use a custom ComfyUI workflow with OpenMontage?▼

Export the workflow from ComfyUI in API format, then pass it as workflow_json or workflow_path along with the output_node parameter. The output_node must be the ID string of the terminal saver node, such as SaveImage or VHS_VideoCombine, not its class name.

Which ComfyUI workflows work on 8GB or 12GB GPUs?▼

Lower-footprint options include Wan 2.1 1.3B, LTXV FP8 or quantized workflows, and Wan 2.2 GGUF community workflows. The bundled Wan 2.2 14B FP8 video workflows target 16GB-class GPUs, so they are not suitable for smaller cards.

Why does my ComfyUI workflow fail to submit?▼

The most common cause is exporting the workflow in UI layout format instead of API format. Re-export it from ComfyUI with API format enabled, and verify the server is reachable at the configured COMFYUI_SERVER_URL.

What happens when a ComfyUI generation times out?▼

A timeout is recoverable: take the prompt_id from the error and pass it back as resume_prompt_id to continue waiting on the same job instead of resubmitting. You can also check ComfyUI history, since the server may have already completed the prompt.

Can I add LoRAs to any ComfyUI workflow?▼

The current tools do not inject LoRAs into arbitrary graphs. Provide a workflow that already contains a LoraLoader or LoraLoaderModelOnly chain, place the LoRA files under ComfyUI/models/loras/, and record each LoRA name with its strength values for provenance.

How does comfyui_music handle prompts and lyrics?▼

The bundled ACE-Step v1 workflow maps prompt to the tags field for style, genre, and mood, while lyrics is a separate optional field supporting [verse]/[chorus] structure tags and language-code prefixes. Leave lyrics empty for instrumental output.

Related Skills

Explore More Design & Creative Skills→
jackspacejackspace
16

comfyui-workflow-helper

Guide ComfyUI workflow creation, node configuration, and optimization.

Community
Advanced
calesthiocalesthio
123

ffmpeg-media-finishing

Audit and finish media into verified delivery files with ffmpeg and ffprobe.

Community
Advanced
ahostbrahostbr

ls-comfy-to-liteimage

Translate ComfyUI workflow JSON node graphs into LiteImage REST API call sequences.

Community
Advanced
calesthiocalesthio
55.1k

minimax-h3

Generate MiniMax Hailuo 3.0 video clips through hosted APIs or local ComfyUI workflows.

Community
Intermediate