# Screens

> Two-way screen sync with the project's design tool. "prompt <flow" writes one wireframe-level design brief per platform (docs/prompts/screens/<project/<NNN-<flow/<platform.md) — always the flow's full screen blueprint, regenerated in place, never a change note, transcribing each screen's contract-pinned components and their rules — commissioning one interactive page per flow per platform on the design tool's canvas under a strict naming contract (pages <flow--<platform, frames named by the pinned screen codes, happy paths stitched into index--<platform); it also maintains each platform canvas project's conventions CLAUDE.md (CLAUDE--<platform.md — one design project per platform, generated sections regenerated, canvas-owned section preserved); the files are the deliverable, never run against the canvas; "import [flow]" reads the designed pages back as data, diffs them against the Screens contracts (components included), folds canvas-discovered conventions back into the conventions file, and routes every accepted contract delta through /vwf:blueprint — this skill never edits a flow doc itself.

## Facts
- Page: https://tashan.sh/capability/skill-virajp-screens
- tashan id: skill:virajp/screens
- Source: https://github.com/virajp/ai-plugins
- Type: skill
- Category: design
- tashan score: not scored (catalogued only — too little public evidence)
- Adoption: 9.0
- Upkeep: 81.0
- Freshness: 98.0
- Evidence coverage: 84% of the inputs this score can use
- Health: active
- Instruction depth: not yet graded
- License: MIT
- Official: no

## Install

```sh
cp -r screens ~/.claude/skills/
```

## Security audit
Not scanned. We audit npm-published capabilities; this one has no npm package we can resolve, or has not reached the queue. This is not a clean bill of health.

---
Measured 2026-08-15 by tashan (https://tashan.sh) from public evidence. Scorer s5.
