StyleseedSAFE

The most comprehensive toolkit for Claude Code -- 135 agents, 35 curated skills, 42 commands, 176+ plugins, 20 hooks, 15 rules, 7 templates, 14 MCP configs, 26 companion apps, 52 ecosystem entries, and more.

The instruction file, verbatim from the audited commit — this is the text the model reads, and the surface the audit's instruction layer examines. Quoted here so you can judge it without cloning anything.

# StyleSeed — Design Judgment Engine

> Teaches Claude Code design judgment, not just design data. 69 visual rules that make AI output look designed, not generated.

## What It Teaches

- **Color discipline** — `#2A2A2A` as refined black, 5-level grayscale, one accent color maximum
- **Spatial rhythm** — alternating section heights, 2:1 number-to-unit ratios
- **Information hierarchy** — card/background separation, progressive density
- **Shadow & elevation** — 4% opacity ceiling, dark mode border substitution
- **Component variance** — never 4 identical KPI cards, stagger content types
- **Motion & feedback** — 200ms normal, spring for entrance, ease-out for exit

## Quick Start

```bash
# Copy engine into your project
cp -r styleseed/engine/* your-project/

# Or just point Claude at the repo
# "Refer to https://github.com/bitjaru/styleseed — read engine/CLAUDE.md and DESIGN-LANGUAGE.md"
```

## Includes

- 69 visual design rules (brand-agnostic)
- 48 React components (shadcn-style, Tailwind CSS v4 + Radix UI)
- 11 slash commands (`/ss-page`, `/ss-review`, `/ss-audit`, `/ss-lint`, etc.)
- Swappable brand skins (Toss, Stripe, Linear, Vercel, Notion, Raycast, Arc)

## Links

- **Repository:** https://github.com/bitjaru/styleseed
- **Live demo:** https://styleseed-demo.vercel.app
- **License:** MIT

SAFEgrade B · trust 89/100 Nothing in the source contradicts what it says it does. Grade A is reserved for packages that have also passed the behavioural sandbox.

What the source does

Filesystem

none-observed

Network

none-observed

Shell

none-observed

Dependencies

pinned

Secrets in source

none-found

Findings (1)

Gates applied: no_behavioural_pass.