vgl

Convert natural language image requests into deterministic VGL JSON prompts for Bria's FIBO models.

63|5|Updated Feb 5, 2026
One-click install
npx skills add https://github.com/Bria-AI/bria-skill --skill vgl
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: vgl
Source: https://github.com/Bria-AI/bria-skill/tree/main/skills/vgl
Command: npx skills add https://github.com/Bria-AI/bria-skill --skill vgl

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

Converts ambiguous natural language image requests into deterministic, structured VGL JSON prompts for Bria's FIBO image generation models, enabling repeatable results.

Core Features & Use Cases

  • Deterministic prompts: Replaces vague requests with explicit objects, lighting, composition, and style.
  • Multi-mode support: Generate, Edit, Edit_with_Mask, Caption, and Refine prompts for generation, editing, inpainting, outpainting, and captioning use cases.
  • Real-world example: Convert "a sunny beach at sunset with palm trees" into a full VGL JSON ready for Bria's API to generate a scene.

Quick Start

Input a natural language scene description and receive a complete VGL JSON prompt ready for Bria's FIBO models.

Frequently Asked Questions about vgl

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I convert natural language prompts into deterministic JSON for image generation?

To convert natural language prompts into deterministic JSON, you input a scene description and the skill outputs a strict VGL JSON schema with explicit objects, lighting, composition, and style for repeatable image generation results.

What is VGL JSON and when do I need it for text-to-image generation?

VGL JSON is a structured prompt format for Bria's FIBO models that replaces vague text requests with deterministic parameters. You need it when you want repeatable, controlled image generation results instead of ambiguous outputs.

Can I use VGL JSON prompts for image editing and inpainting tasks?

Yes, VGL JSON supports multiple modes including Generate, Edit, Edit_with_Mask, Caption, and Refine, covering text-to-image generation, image editing, inpainting, outpainting, and background generation tasks.

How do I create structured prompts from natural language for Bria FIBO models?

You provide a natural language scene description like a sunny beach at sunset, and the skill converts it into a complete VGL JSON prompt with explicit composition, lighting, and style details ready for Bria's FIBO API.

Does this deterministic prompt approach support captioning and outpainting workflows?

Yes, the skill supports Caption mode for image captioning and Refine mode for outpainting, alongside Generate, Edit, and Edit_with_Mask modes, all outputting strict VGL JSON schemas for Bria's FIBO models.