dubery-v3-pipeline

Automate product image generation from category selection to final render.

Updated Mar 9, 2026
One-click install
npx skills add https://github.com/RASCLAW/DuberyMNL --skill dubery-v3-pipeline
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: dubery-v3-pipeline
Source: https://github.com/RASCLAW/DuberyMNL/tree/main/.claude/skills-archive-v1/dubery-v3-pipeline
Command: npx skills add https://github.com/RASCLAW/DuberyMNL --skill dubery-v3-pipeline

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

End-to-end automation for generating and validating product imagery, dramatically reducing manual setup and iteration time.

Core Features & Use Cases

  • Coordinates category selection, prodref and sidecar loading, spec filtering, scene randomization, validation, and image generation for consistent, production-ready visuals.
  • One-image-at-a-time workflow with strict fidelity prompts and state tracking to minimize hallucinations and errors.
  • Supports multiple UGC categories and a repeatable pipeline with per-image provenance and layout history.

Quick Start

Select a category and run the v3 pipeline to generate a single image from prodref load through sidecar application, detail filtering, scene randomization, prompt validation, and final render.

Frequently Asked Questions about dubery-v3-pipeline

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

FAQPage Schema
How do I automate end-to-end product image generation from a category prompt to final render?

End-to-end product image generation is automated by coordinating category selection, prodref loading, scene randomization, and fidelity-prompt construction to render production-ready visuals. The pipeline applies sidecar-based detail filtering and per-image validation to minimize hallucinations.

What is the best way to maintain consistent product imagery across multiple UGC categories?

Consistent product imagery across UGC categories is maintained through a repeatable pipeline that tracks per-image provenance and layout history. The workflow generates one image at a time using strict fidelity prompts and state tracking to ensure uniform quality.

How does deterministic scene randomization work for product content generation?

Deterministic scene randomization for product content generation applies step-wise scripts to vary visual layouts while maintaining reproducibility. This process pairs with fidelity-prompt construction to ensure generated scenes remain accurate to the loaded prodref specifications.

Can I use sidecar files for detail filtering when generating product visuals?

Sidecar files are actively used for detail filtering during the product visual generation pipeline. The workflow loads sidecar data alongside category-driven prodrefs to apply specific visual constraints and validate image fidelity before final render.

Why does my product image generation workflow require per-image validation checks?

Per-image validation checks are required in product image generation to minimize hallucinations and output errors before finalizing the render. The pipeline enforces strict fidelity prompt validation and state tracking to guarantee production-ready visuals.

Are there limitations to generating multiple product images simultaneously in an automated workflow?

Automated product image generation operates strictly as a one-image-at-a-time workflow to maintain strict fidelity prompts and accurate state tracking. This sequential processing prevents hallucinations but limits batch generation speed for high-volume output.