visual-ralph - Frontend UI visual implementation with pixel-level verification
Orchestrate frontend UI implementation from visual references using pixel-level verdict and iteration
Tags
Updated: 2026-05-11Capabilities
Typical Inputs
Typical Outputs
What this skill does
- generate UI mockup reference
- capture URL baseline
- run visual verdict
- generate pixel diff
- extract design tokens
- compare screenshots
- validate implementation
Inputs
- User UI description
- Live URL
- Static reference image
- Target repository
- Viewport specifications
- Content state notes
- Interaction parity notes
- Screenshot command
Outputs
- Approved reference image
- Implemented frontend UI
- Visual verdict score
- Pixel diff evidence
- Design system tokens
- Component library
- Screenshot artifacts
Requirements
- Target repository
- $ralph skill
- Vision agent access
- Image generation capability
- Screenshot capture tool