explaingit

16nic/comfyui-agnes-ai

19PythonAudience · vibe coderComplexity · 2/5LicenseSetup · moderate

TLDR

A ComfyUI plugin that adds seven workflow blocks connecting to the Agnes AI service, covering text chat, image analysis, image generation up to 4K, image editing, and video generation from text or images.

Mindmap

mindmap
  root((comfyui-agnes-ai))
    What it does
      Agnes AI integration
      7 workflow blocks
      Auto retry on errors
    Block types
      API key config
      Text chat
      Image to description
      Text to image
      Image editing
      Text to video
      Images to video
    Image generation
      Up to 4K resolution
      11 aspect ratios
    Setup
      Clone to custom nodes
      Install requirements
      Set API key once
Click or tap to explore — scroll the page freely

Code map

Detail Auto

An interactive map of this repo's files and how they connect — its source is parsed live in your browser. Click Visualize to build it.

filefunction / class

Things people build with this

USE CASE 1

Build a ComfyUI workflow that turns a text prompt into a 4K image using Agnes AI without writing any code.

USE CASE 2

Add an image-to-video block to your existing ComfyUI pipeline to automatically generate short videos from still images via Agnes AI.

USE CASE 3

Analyze an existing image with Agnes AI to get a text description you can feed into another generation block in your workflow.

Tech stack

PythonComfyUI

Getting it running

Difficulty · moderate Time to first run · 30min

Requires a free Agnes AI API key and cloning into ComfyUI's custom_nodes folder before restarting ComfyUI.

MIT License, use freely for any purpose including commercial use, as long as you keep the copyright notice.

In plain English

This is a plugin for ComfyUI, a visual tool that lets people build image and video generation workflows by connecting blocks together on a canvas. The plugin adds seven new blocks, each of which sends a request to the Agnes AI service and returns the result back into your workflow. The seven blocks cover a range of tasks: saving your API key so you do not have to retype it, text-based chat with an AI, analyzing an image to produce a text description that could recreate it, turning a text description into an image, editing an existing image based on text instructions, turning a text description into a short video, and turning one or more images into a short video. Image generation supports resolutions up to 4K and eleven aspect ratios. Video generation is handled asynchronously and the README notes it typically takes two to six minutes to complete. To use this plugin you need a free API key from the Agnes AI platform. The recommended setup is to run the API Key Config block once after installation, which saves the key to a local file so all other blocks can find it automatically. The key is stored as plain text on your own machine and is not displayed visibly in the interface. Installation follows the standard ComfyUI plugin pattern: clone the repository into the custom_nodes folder, install the Python dependencies listed in requirements.txt, and restart ComfyUI. The plugin includes automatic retry logic for server errors, so temporary outages on the Agnes AI side should recover without manual intervention. The README is written in Chinese. The project is released under the MIT License.

Copy-paste prompts

Prompt 1
Help me set up the comfyui-agnes-ai plugin in my ComfyUI custom_nodes folder and configure the API Key Config block so all other Agnes AI blocks find my key automatically.
Prompt 2
I want to build a ComfyUI workflow using comfyui-agnes-ai that takes a text prompt, generates a 1080p image, then turns that image into a short video. Show me how to connect the blocks.
Prompt 3
My Agnes AI video generation block in ComfyUI is taking longer than 6 minutes. How does the async polling work and when should I expect a timeout or error?
Prompt 4
I want to use the image-to-description block in comfyui-agnes-ai to analyze a photo and then feed that description into a text-to-image block. Walk me through wiring this workflow.
Open on GitHub → Explain another repo

← 16nic on gitmyhub — every repo by this author, as a profile.

Verify against the repo before relying on details.