vverify

Compare screenshots to reference images and generate JSON scorecards.

2|1|Updated Jul 9, 2026
One-click install
npx skills add https://github.com/yy1588133/oh-my-snow --skill vverify
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: vverify
Source: https://github.com/yy1588133/oh-my-snow/tree/main/assets/skills/oms/vverify
Command: npx skills add https://github.com/yy1588133/oh-my-snow --skill vverify

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) components.

What problem does it solve?

Vverify addresses the challenge of confirming visual consistency for tasks requiring a visual check, such as ensuring UI components match designs or ensuring UI updates do not introduce issues.

Core Features & Use Cases

  • Screenshot Comparison: Automatically compare a current screenshot to a reference image for visual consistency.
  • Scorecard Generation: Generates a scorecard with a JSON output detailing the level of visual match between the reference and the current image.
  • Iteration Loop Triggering: Repeatedly triggers the visual check until a certain score threshold is reached, typically 90.
  • Fallback for Image Unavailability: Offers a fallback option to describe visual discrepancies when images are not available.

Quick Start

Trigger the vverify skill with the current and reference screenshots.

Frequently Asked Questions about vverify

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

FAQPage Schema
How do I automate screenshot comparison for UI consistency checks?

Screenshot comparison automates visual QA by comparing current screenshots against reference images to generate a detailed JSON scorecard. It validates UI components and maintains layout consistency without manual inspection.

Can I trigger automated visual QA repeatedly until a layout match score is reached?

Automated visual QA supports an iteration loop that repeatedly triggers screenshot comparison until a specific score threshold, typically 90, is achieved. This ensures UI updates meet visual consistency requirements before proceeding.

What is the best way to generate a visual scorecard for UI verification?

Generating a visual scorecard involves using Python-based image comparison techniques to compare a current screenshot to a reference image. The output is a detailed JSON file indicating the level of visual match for validation.

Do I need Python image comparison libraries to perform screenshot-based visual QA?

Yes, screenshot-based visual QA requires access to Python image comparison libraries to handle image data inputs and process the visual component comparison. This environment setup is necessary to automate the validation scorecard generation.

How does visual verification handle UI checks when reference images are unavailable?

Visual verification offers a fallback option to describe visual discrepancies when reference images are not available. This ensures the visual consistency check can still proceed by detailing layout issues textually.