nregret/Comfyui-Anima-Tools

What it solves

Anima-Tools addresses the difficulty of manually searching for and formatting complex prompts for anime-style AI art. It eliminates the need to memorize thousands of artist names, character traits, or clothing tags by providing a visual, searchable interface directly within ComfyUI, while also automating the random combination of these elements for creative exploration.

How it works

The project implements a set of custom ComfyUI nodes and frontend panels that integrate large datasets of anime artists, characters, and clothing. It uses JavaScript-based visual selectors to let users pick styles and tags from galleries, which are then passed to Python backend nodes. It also includes a "Prompt Composer" that logically combines these elements (Artist $\rightarrow$ Character $\rightarrow$ Clothing $\rightarrow$ Background $\rightarrow$ Pose) using specific rules to avoid conflicting tags (e.g., ensuring a character doesn't wear two pairs of shoes).

Who it’s for

Digital artists and AI creators using ComfyUI who specialize in anime/manga aesthetics and want a more intuitive, visual way to manage prompts and LoRAs.

Highlights

• Massive Databases: Access to over 40,000 Danbooru artists and extensive character/clothing libraries.
• Visual Selectors: Interactive panels with 3:4 ratio preview cards and multi-dimensional filtering (e.g., by hair color, eye color, or popularity).
• Anima Prompt Composer: A node that automatically generates random, logically consistent prompt combinations with separate seeds for characters and clothing.
• Integrated LoRA Management: A dedicated loader panel for searching, previewing, and downloading LoRAs from Civitai.
• Smart Formatting: Automatically handles tag normalization (e.g., adding @ prefixes for artists) and prevents punctuation errors like double commas.