visual-prior-art-survey

Surveys documented UI and interaction conventions from design systems, WCAG, ARIA APG, and platform guidelines.

1|1|Updated May 24, 2026
One-click install
npx skills add https://github.com/bm629/agent-skills --skill visual-prior-art-survey-bm629
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: visual-prior-art-survey
Source: https://github.com/bm629/agent-skills/tree/main/skills/visual-prior-art-survey
Command: npx skills add https://github.com/bm629/agent-skills --skill visual-prior-art-survey-bm629

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires pyyaml, pytest, jsonschema, and includes scripts (resource) and references (resource) components.

What problem does it solve? Before any wireframe or design system is produced, teams need to know what visual and interaction conventions the industry has already documented and what accessibility criteria actually bind. This Skill mines that documented prior art from governed sources and produces schema-validated artifacts a downstream design skill can consume, instead of relying on an agent's recall or screenshot galleries. ## Core Features & Use Cases - Vocabulary map derivation: Builds a UI-pattern vocabulary map across five axes (component, pattern, screen-archetype, platform-context, design-system) with typed expansions and honest provenance. - Angle-based search execution: Runs one named search angle across design-system documentation, the ARIA Authoring Practices Guide, WCAG success criteria, platform HIGs, and the deceptive-pattern corpus, recording every query verbatim with a coverage grid. - Deep-read extraction and synthesis: Deep-reads one convention source into a structured record carrying DTCG design tokens verbatim, then synthesizes a convention register and seven-section report. - Use Case: Before designing a consumer marketplace app, run the survey to extract the WCAG 2.2 AA criteria, ARIA interaction contracts, and relevant design-system conventions that bind the project, then hand the register to the downstream design-system authoring skill. ## Quick Start Use visual-prior-art-survey to build the UI-pattern vocabulary map for my product scope before any wireframing begins.

Frequently Asked Questions about visual-prior-art-survey

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

FAQPage Schema
How do I research UI design conventions before wireframing?

Run the survey's four procedures: derive a UI-pattern vocabulary map from the scope, execute search angles across design systems and normative specifications, deep-read convention sources into records, then synthesize a convention register and report. Every artifact is validated by a deterministic gate.

What sources does a visual prior-art survey cover?

It covers governed design systems (Material, Carbon, Polaris, Primer, Fluent, GOV.UK), the ARIA Authoring Practices Guide, WCAG 2.2 success criteria, platform HIGs (Apple, Android, Windows), and the deceptive.design corpus. Screenshot galleries like Mobbin and Dribbble are excluded by terms and modality.

Why does the survey exclude screenshot galleries like Mobbin and Dribbble?

Screenshot galleries are excluded on two independent grounds: their subscription terms forbid automated extraction, and a screenshot cannot be extracted into the textual, schema-validated artifact that downstream design skills consume. Only documented conventions with rationale are extractable.

Does the survey report what shipped products actually do?

No. It reports what documented systems prescribe, never what shipped products actually do, because adoption and divergence data would require a screenshot corpus it deliberately does not use. Read its output as documented convention, not as the screens a product needs.

How are design tokens handled in the survey output?

Design-system records carry tokens as a fenced DTCG block in the record body, copied verbatim into the convention register. Tokens are never merged across systems, and any claimed token format must state its version, such as DTCG 2025.10.

What happens when a search finds no documented convention?

A zero-hit result is recorded as a receipt that the search ran, with every query verbatim. This is kept distinct from unreachable sources (typed failures) and forbidden sources (policy decisions), so absence of evidence is never confused with evidence of absence.