visual-verdict

Compare UI screenshots against reference images and return a JSON verdict.

Updated Jan 4, 2026
One-click install
npx skills add https://github.com/YeaChan05/Remittance --skill visual-verdict-yeachan05
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: visual-verdict
Source: https://github.com/YeaChan05/Remittance/tree/main/.codex/skills/visual-verdict
Command: npx skills add https://github.com/YeaChan05/Remittance --skill visual-verdict-yeachan05

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This description provides automated visual QA by comparing a generated UI screenshot against one or more reference images and producing a deterministic JSON verdict that guides design iterations.

Core Features & Use Cases

  • Pixel-perfect comparison across multiple references and optional category hints.
  • Deterministic JSON verdict including score, verdict, category_match, differences, suggestions, and reasoning.
  • Use Case: regression visual checks after UI changes or theme updates.

Quick Start

Provide a generated_screenshot and one or more reference_images, then run the visual-verdict check to obtain a JSON verdict.

Frequently Asked Questions about visual-verdict

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

FAQPage Schema
How do I compare a UI screenshot against a reference image for visual regression testing?

Visual regression testing for UI screenshots involves comparing a generated screenshot against one or more reference images to determine pixel-accuracy. This Skill outputs a deterministic JSON verdict with fields like score, verdict, and differences to guide design iterations.

What is pixel-diff visual verification for design reviews?

Pixel-diff visual verification is an automated visual QA process that compares UI screenshots against reference images to produce a strict JSON verdict. It returns fields including category_match, suggestions, and reasoning to support design reviews and regression checks.

Can I use visual verification to check UI changes across both web and mobile apps?

Yes, visual verification applies to visual QA tasks across web and mobile apps. By providing a generated screenshot and reference images, you receive a JSON verdict containing a score, verdict, and identified differences to validate UI consistency after changes.

What's the best way to get structured JSON output from a screenshot comparison?

To get structured JSON output from a screenshot comparison, provide a generated screenshot and one or more reference images to the visual-verdict check. It returns a deterministic JSON object with score, verdict, category_match, differences, suggestions, and reasoning fields.

Does visual QA comparison support category hints for matching specific UI components?

Yes, visual QA comparison supports an optional category_hint parameter. Providing this hint alongside your generated screenshot and reference images helps refine the pixel-accuracy verdict and the resulting JSON output for targeted design reviews.

Why does my visual regression check return a low score despite minor UI differences?

A visual regression check returns a low score when pixel differences are detected between the generated screenshot and reference images. The JSON verdict includes a differences field detailing discrepancies, along with suggestions and reasoning to guide necessary design iterations.