Screenshot Surgery — Taste-Skills — AI skill

When to use it

Provide a screenshot or rendered interface to a vision-capable agent. Useful for teaching visual observation and documenting why a screen feels disorganized. Deliberately separates observed hierarchy from recommendations; it is not a coded annotation tool or an accessibility audit. Intentional versus accidental tension requires context. Upstream labels the project experimental and requires human review. Upstream reviewed only; not installed or runtime-tested. GitHub stars are repository-wide, not skill-specific.

Compatible agents

Claude CodeCursorWindsurfOpenCodeClineCodexAiderContinue

Installation

Manual setup: download skills/perception/screenshot-surgery/SKILL.md from https://github.com/Dragoon0x/taste-skills into .taste-skills/screenshot-surgery/SKILL.md in your project and point your agent at that directory, as described by the upstream README. Ask it to apply screenshot-surgery to your supplied design. The README CLI examples currently reference 0xDragoon/taste-skills, which returned GitHub 404 when checked; use the verified current source instead.

Review the source and instructions before installing in your agent.

License

MIT

Last checked Oct 18, 2026.

Pair this skill with a visual reference

A skill guides how your agent works; a design reference shows the direction you want. Choose an example for a specific job, such as the headline hierarchy or form layout, and keep your project's existing design rules explicit.

Start with design references you can inspect, then use our guide to briefing an AI coding agent to turn observations into requirements. If you need implementation components, compare UI libraries for your stack.

These pages are public. Browsing a skill does not install it, and TastefulKit's design-reference MCP tools do not search the AI-skill directory.