AI Tagger for Anchorpoint

This project is an AI-based asset tagging tool for Anchorpoint

It uses OpenAI's GPT-4o model to automatically generate tags for assets based on their preview and for folders based on the hierarchy of their contents.

Installation

• Open Workspace Settings
• Open Actions tab
• Press Import button
• Paste https://github.com/Hermesiss/anchorpoint-action-openai-tagger into the Repository URL field
• Press Connect repository button

Configuration

You need to provide your OpenAI API key in the Action settings.

Obtain one from OpenAI

You need at least $10 in your account to use the API.

• Open Workspace Settings -> Actions -> Ai Folder and File Tools -> Settings
• Paste your API key into the OpenAI API Key field
• Press Apply

Usage

Tagging folders

This action will automatically tag folders based on the hierarchy of their contents and create 3 attributes: AI-Engines, AI-Types and AI-Genres.

• Change to the List View (Ctrl+2)
• Select one or more folders
• Right-click on any of the selected folders
• Select Tag Folders with AI
• Wait for the preparation stage to finish
• You will be prompted with token count and cost estimation and a confirmation dialog
• If you confirm, the action will start, and you will be notified when it finishes

Tagging assets

This action will automatically tag assets based on their preview and create 3 attributes: AI-Types, AI-Genres and AI-Objects.

• Change to the List View (Ctrl+2)
• Select one or more assets - images, models, materials. All that have a preview
• Right-click on any of the selected assets
• Select Tag Files with AI
• Wait for the preparation stage to finish - generating previews and optimizing them to reduce the token count
• You will be prompted with token count and cost estimation and a confirmation dialog
• If you confirm, the action will start, and you will be notified when it finishes

---