Save 4 hours! We auto-setup your workflow! Free!

Drop your workflow.json — we handle every dependency, custom node, and model. Just open the link and run.

Auto-Setup Workflow Json (Free) Now!
ComfyUI > Nodes > jk-comfyui-helpers

ComfyUI Extension: jk-comfyui-helpers

Repo Name

jk-comfyui-helpers

Author
kostenickj (Account age: 3923 days)
Nodes
View all nodes(1)
Latest Updated
2024-12-19
Github Stars
0.01K

How to Install jk-comfyui-helpers

Install this extension via the ComfyUI Manager by searching for jk-comfyui-helpers
  • 1. Click the Manager button in the main menu
  • 2. Select Custom Nodes Manager button
  • 3. Enter jk-comfyui-helpers in the search bar
After installation, click the Restart button to restart ComfyUI. Then, manually refresh your browser to clear the cache and access the updated list of nodes.

Visit ComfyUI Online for ready-to-use ComfyUI environment

  • Free trial available
  • 16GB VRAM to 80GB VRAM GPU machines
  • 400+ preloaded models/nodes
  • Freedom to upload custom models/nodes
  • 200+ ready-to-run workflows
  • 100% private workspace with up to 200GB storage
  • Dedicated Support

Run ComfyUI Online

jk-comfyui-helpers Description

jk-comfyui-helpers adds a detachable image feed, prompt autocomplete for LoRAs, wildcards, and tags, and a single high-resolution fix node with latent noise and upscaling for ComfyUI.

jk-comfyui-helpers Introduction

jk-comfyui-helpers is a collection of visual helpers and nodes for ComfyUI. It helps you keep an eye on generated images, enter prompts more easily, and build a high-resolution workflow with fewer separate steps.

The extension has three main parts:

  • Custom Feed Window: View generated images in a separate window, including on another monitor.
  • Custom Autocomplete: Get suggestions while writing prompts, including LoRAs, wildcards, and words from tag files.
  • Easy Hi Res Fix Node: Use a single node for a high-resolution process inspired by A1111 and Forge.

How jk-comfyui-helpers Works

Think of the feed window as a separate viewing desk for your images. Your ComfyUI workflow can stay on one screen while the image feed sits on another. You can switch between a grid of images and a single-image view, and choose which nodes contribute images to the feed.

Autocomplete helps as you type in a prompt editor. It can suggest entries from available sources so you do not have to remember every tag, wildcard, or LoRA name. For a LoRA you use often, you can save preferred activation text and a weight; selecting that LoRA through autocomplete can then insert those preferences for you.

The Easy Hi Res Fix Node brings several parts of a high-resolution process into one node. It adds a small amount of noise to the image’s latent representation before upscaling, following the approach the author describes for A1111 and Forge. By default, it generates that noise on the GPU.

jk-comfyui-helpers Features

Custom Feed Window

The custom feed window can be moved away from the main ComfyUI interface. This is especially useful with two monitors: you can work on your node graph on one screen and review image results on the other.

Opening the custom feed window

You can choose the view that suits what you are doing:

  • Grid mode lets you scan multiple results together, useful when comparing variations.
  • Single-image mode gives one result more space, useful when checking details.

Switching between grid and single-image modes

You can also toggle output visibility for individual nodes. For example, if a workflow produces both intermediate previews and final images, you can hide outputs you do not want in the feed.

Choosing which node outputs appear in the feed

The feed includes a JSON prompt explorer for PNG images. It provides a way to inspect prompt information associated with a supported PNG result. The original README specifies PNG support only.

Exploring prompt information in a PNG image

Custom Autocomplete

The autocomplete is the author’s customized version of the feature from ComfyUI-Custom-Scripts. It is intended to make ComfyUI prompt editing feel more like the tag autocomplete experience available in Forge and A1111.

Autocomplete suggestions in a prompt editor

LoRA preferences let you save the activation text and weight you prefer for a particular LoRA. This can make repeated prompting more consistent. For example, if you normally pair a LoRA with specific trigger words, you can store those words instead of typing them each time.

To edit a LoRA’s preferences, open its dialog by either right-clicking the LoRA in the ComfyUI sidebar or clicking the information icon in its autocomplete entry. The dialog includes additional fields for saving those preferences.

Editing LoRA autocomplete preferences

Autocomplete also supports wildcards and custom word files. The author notes that it may detect wildcards installed through other extensions that support them, but does not confirm compatibility with every extension. You can add your own text or CSV word files to the extension’s tags directory. A danbooru.csv file is included by default; you can remove it if you do not want suggestions from that file.

Wildcard autocomplete suggestions

LoRAs mentioned in a prompt need a prompt processor to take effect. The author recommends the prompt-processing nodes in the ComfyUI Inspire Pack. Autocomplete helps you enter the text, but it does not replace that processing step.

Easy Hi Res Fix Node

The Easy Hi Res Fix Node is the author’s attempt to reproduce the A1111/Forge high-resolution fix process in a single ComfyUI node. It adds a small amount of noise to the latent before upscaling and generates that noise on the GPU by default.

Easy Hi Res Fix Node

The node requires both the ComfyUI Inspire Pack and ComfyUI Noise. If either is missing, the high-resolution workflow will not have all the components it needs.

Troubleshooting jk-comfyui-helpers

Autocomplete suggestions conflict or behave unexpectedly

  1. Check whether you also have the original autocomplete from ComfyUI-Custom-Scripts enabled.
  2. Open the ComfyUI settings manager.
  3. Disable one of the two autocomplete implementations. The author notes that they conflict when both are active.

A LoRA appears in autocomplete but does not affect the generated image

  1. Confirm that your workflow includes a processor that handles LoRA references written in prompts.
  2. Check the Inspire Pack prompt-processing nodes, which the author recommends for this purpose.
  3. Check that the LoRA text and saved preferences are what you intended to use.

The Easy Hi Res Fix Node is unavailable or its workflow is missing components

  1. Check whether ComfyUI Inspire Pack is available in your ComfyUI setup.
  2. Check whether ComfyUI Noise is available as well.
  3. Make sure both dependencies are active before using the node.

The JSON prompt explorer does not work with an image

Check the image format first. The explorer is described as supporting PNG files only; support for other formats is not documented.

Can I remove the included Danbooru suggestions?

Yes. The author says you can delete the included danbooru.csv file from the extension’s tags directory if you do not want its entries in autocomplete.

Learn More about jk-comfyui-helpers

jk-comfyui-helpers Related Nodes

RunComfy
Copyright 2025 RunComfy. All Rights Reserved.

RunComfy is the premier ComfyUI platform, offering ComfyUI online environment and services, along with ComfyUI workflows featuring stunning visuals. RunComfy also provides AI Models, enabling artists to harness the latest AI tools to create incredible art.

jk-comfyui-helpers detailed guide | ComfyUI