skills.rest
TylerSimons1127
TylerSimons1127COMMUNITY
Explore all skills by this author →
GitHub

comfy-relay

Generates images on a remote ComfyUI GPU machine through a local LAN relay.

Updated Aug 21, 2026
Search Tags:#image generation#gpu#relay#LAN#comfyui#txt2img#img2img
One-click install
npx skills add https://github.com/TylerSimons1127/vibe --skill comfy-relay-tylersimons1127
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: comfy-relay
Source: https://github.com/TylerSimons1127/vibe/tree/main/skills/creative/comfy-relay
Command: npx skills add https://github.com/TylerSimons1127/vibe --skill comfy-relay-tylersimons1127

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve? Your laptop has no GPU, but a ComfyUI instance runs on a gaming PC elsewhere on your LAN. This Skill lets you queue text-to-image, image-to-image, and video generations on that remote machine through a local relay, without Tailscale or manual file transfers. ## Core Features & Use Cases - Remote generation via local relay: A relay on 127.0.0.1:7899 proxies prompts to the Comfy PC, polls /history, fetches output images, and returns base64 in one call. - Workflow templates: Ships API-format graphs for txt2img, img2img, and Wan2.2 text-to-image, plus UI-format graphs that render visually in the ComfyUI canvas. - Output gallery and verification: Serves a live thumbnail gallery at /output, auto-saves images to a local folder, and validates that generated files are real images before reporting success. - Use Case: Ask the agent to generate "a fox in snow, photoreal" — it checks relay health, submits the txt2img workflow with your prompt/steps/seed, saves the PNG locally, and delivers it back in chat. ## Quick Start Start the relay with COMFY_URL pointing at the Comfy PC, then ask the agent to generate an image from a text prompt using the txt2img workflow.

Frequently Asked Questions about comfy-relay

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

FAQPage Schema
How do I run ComfyUI generation from a laptop without a GPU?▼

Run the comfy_relay.py server locally and point COMFY_URL at the ComfyUI machine on your LAN. Then call comfy_client.py generate with a workflow name and prompt parameters; the relay queues the job, polls history, and returns the images.

How do I find the ComfyUI machine's current IP address on my LAN?▼

Read the installations.json file in the Comfy Desktop AppData folder and take the remoteUrl of the entry named "ComfyUI (1)". Do not hardcode the IP because LAN DHCP can reassign it.

Why does the ComfyUI model-listing API return an empty array?▼

On this ComfyUI build, endpoints like /api/models/checkpoints return [] even though model files exist on disk. Get real filenames from the node dropdowns in the ComfyUI interface or by listing the models/ folders directly on the Comfy PC.

Why does my workflow JSON not render in the ComfyUI canvas?▼

API-format graphs use node-id keys without positions and are for server execution only. To view a graph visually, export or build a UI-format file with node pos/size/title fields and a links array, then load it in the editor.

What should I do when the relay port 7899 stops responding?▼

A killed relay can leave a Python child process still bound to port 7899. Run netstat to find the PID holding the port, taskkill it, then restart the relay so requests do not silently hit the stale instance.

Can I add a new ComfyUI workflow without editing files?▼

Yes, POST the exported API-format graph to the relay's /upload_workflow endpoint with a name and graph body. The graph should include a _params map linking logical names to node IDs and fields.

Related Skills

Explore More Design & Creative Skills→
jackspacejackspace
16

comfyui-workflow-helper

Guide ComfyUI workflow creation, node configuration, and optimization.

Community
Advanced
x402agentx402agent

openai-image-gen

Generate batch images from prompts via the OpenAI Images API into an index.html gallery.

Community
Intermediate
dawsonblockdawsonblock

cli-anything-comfyui

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

Community
Intermediate
frogrfrogr

austn-tools

Automate TTS and image generation through local AI services.

Community
Intermediate