visual-verdict

Compare generated screenshots to reference images and output a deterministic JSON verdict.

1|Updated Mar 27, 2026
One-click install
npx skills add https://github.com/oneforce/oh-my-gemini-cli --skill visual-verdict-oneforce
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: visual-verdict
Source: https://github.com/oneforce/oh-my-gemini-cli/tree/main/skills/visual-verdict
Command: npx skills add https://github.com/oneforce/oh-my-gemini-cli --skill visual-verdict-oneforce

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Structured, deterministic evaluation of UI screenshots against references to drive design iterations and QA workflows.

Core Features & Use Cases

  • Produce a JSON verdict with score, verdict, category_match, differences, suggestions, and reasoning.
  • Compare a generated screenshot to one or more reference images, returning concrete visual diffs for targeted edits.
  • Use in design QA pipelines to enforce consistent UI fidelity across teams and releases.

Quick Start

Provide a generated_screenshot and one or more reference_images to produce a deterministic JSON verdict for UI comparisons.

Frequently Asked Questions about visual-verdict

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

FAQPage Schema
How do I identify visual differences between a UI screenshot and a reference image?

To identify visual differences between a UI screenshot and a reference image, provide both images to generate a deterministic JSON verdict. This output details the score, category match, and specific differences for design QA.

Can I compare one screenshot against multiple reference images for UI testing?

Yes, you can compare one generated screenshot against multiple reference images simultaneously. The comparison evaluates UI layouts, typography, color, and spacing to return concrete visual diffs.

What is a deterministic JSON verdict for UI screenshots?

A deterministic JSON verdict is a structured output that consistently evaluates UI screenshots against references. It includes a score, verdict, category match, array of differences, suggestions, and concise reasoning.

How do I use visual diff output to drive design iterations?

You can use the visual diff output to drive design iterations by parsing the returned JSON for targeted edits. The structured differences and suggestions guide corrections in typography, color, and spacing across QA pipelines.

Does UI visual diffing work for evaluating layout and spacing inconsistencies?

Yes, UI visual diffing works effectively for evaluating layout and spacing inconsistencies. It analyzes these specific visual categories alongside typography and color to produce a structured comparison verdict.