Overview

The Visualizer agent handles image generation for characters and environments in your scenes. It provides a comprehensive system for creating, editing, and managing visual content through the Visual Library.

The Visualizer agent supports multiple image generation backends, allowing you to choose the best tool for your needs. It can perform three main types of operations:

Text to Image: Generate new images from text prompts
Image Editing: Modify existing images using reference images
Image Analysis: Analyze images using AI to extract information

Key Features

Multiple Backend Support: Works with various image generation services including Google, ComfyUI, AUTOMATIC1111, OpenAI, and more
Style Templates: Configure different visual styles for different types of content (character cards, portraits, scene backgrounds, etc.)
Visual Library Integration: Generated images are managed through the Visual Library, where you can organize, iterate, and save visual assets
Automatic Generation: Optionally allow the agent to automatically generate visual content based on scene context
Prompt Generation: Supports both direct prompts and natural language instructions that incorporate character and scene context

Configuration

The Visualizer agent can be configured through its settings panel, where you can:

Select backends for different operations (text-to-image, image editing, image analysis)
Configure generation timeouts
Set up automatic backend setup and generation
Customize visual styles for different content types

See the Settings page for detailed configuration options.

Backend Documentation

The Visualizer agent supports multiple backends, each with its own configuration requirements:

ComfyUI: Advanced node-based workflow system for image generation and editing
Google: Google's Gemini image models for generation, editing, and analysis
OpenAI: OpenAI's DALL·E 3 and GPT-Image models
OpenRouter: Access multiple AI providers through OpenRouter's unified API
SD.Next: Improved fork of AUTOMATIC1111 with better performance and features
AUTOMATIC1111: Legacy Stable Diffusion WebUI backend (deprecated, use SD.Next instead)

Usage

The Visualizer agent can be accessed through several methods:

Visual Library

The primary interface for image generation is the Visual Library, accessible from the toolbar when a scene is active. You can generate images using either:

Prompt Mode: Direct text descriptions of what you want to create
Instruct Mode: Natural language instructions that incorporate character and scene context

For more information on using the Visual Library, see the Visual Library documentation.

Scenario Tools Shortcuts

Quick shortcuts are available through the scenario tools menu, allowing you to quickly generate images for:

Visualize Scene: Generate images of the current scene environment
Visualize Character: Generate character portraits or cards

These shortcuts support keyboard modifiers: hold ALT to generate prompts only (without creating images), or hold CTRL to use instruction mode.

Director Chat

The Director agent can also generate images through director chat. When chatting with the director, you can ask it to create visual assets for scenes or characters, and it will use the Visualizer agent to generate the requested images.

For more information on director chat, see the Director Chat documentation.

3.7 KiB Raw Blame History