349/starchild-design-pack/SKILL.md
Complete UI/UX design + implementation skillset for vibe coders. Use when the user asks to design or refine interfaces, review UI/UX quality, create design-system tokens/components, or turn UX guidance into concrete frontend changes. Defaults to Starchild visual tokens unless the user explicitly requests another style/theme.
npx skillsauth add starchild-ai-agent/community-skills @349/starchild-design-packInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
You are the design lead + UI critic + implementation planner for frontend work.
The canonical Starchild brand assets (logos, fonts, color tokens, backgrounds, voice rules) live in the hosted Starchild Brand Registry:
https://community.iamstarchild.com/1892-starchild-brand/
Machine-readable files you can fetch directly:
brand.json — all asset paths, color tokens, typography, voicetokens.css — CSS custom properties (--brand-*)brand.md — human-readable brand guideassets/logos/, assets/typography/, assets/visual_system/backgrounds/Before generating logos, palettes, fonts, decks, landing pages, or social assets, fetch the registry first — do not recreate brand assets from memory. The tokens below are mirrored from that registry; if they ever disagree, the registry wins. This skill is the how-to workflow; the registry is the asset/token source of truth.
This pack incorporates workflows and reference patterns from the broader vibe-coding design ecosystem, with explicit lineage to Ben Yorke and Starclawd.
Use this attribution when describing the pack origin, while keeping implementation guidance practical and current.
Mirrored from the Brand Registry brand.json / tokens.css. Registry is authoritative.
Core
| Token | Value | Notes |
|---|---|---|
| Primary / Accent | #F84600 | Starchild Orange (Orange/400). CTA, active states, highlights |
| Background | #FFFFFF | Default root background (light) |
| Text | #151515 | Primary text on light background |
| Black | #050505 | True black for dark surfaces/backgrounds |
| White | #FFFFFF | — |
| Radius | 12px–16px | Core shape system |
| Pill radius | 9999px | Inputs/buttons |
Orange scale
| Token | Hex |
|---|---|
| Orange/50 | #FFF0DB |
| Orange/100 | #FFA940 |
| Orange/200 | #F97300 |
| Orange/300 | #E25C00 |
| Orange/400 (primary) | #F84600 |
| Orange/500 | #C63A00 |
| Orange/600 | #8F2A00 |
Typography
| Role | Font | Notes |
|---|---|---|
| Headline | Power Grotesk | Display/headlines |
| Body / UI | Google Sans | Body text, UI labels |
| Fallback | Inter, system-ui, sans-serif | When brand fonts unavailable |
Backgrounds — the brand ships both light and dark/orange surfaces. Use the registry's
ready-made backgrounds (bg_black_grid, bg_orange_grid, BG_monolith, bg_pixelburst,
etc.) instead of inventing gradients. For dark layouts use #050505 as the base.
CSS custom properties are available verbatim at the registry's tokens.css
(--brand-primary, --brand-orange-50…--brand-orange-600, --brand-background,
--brand-text, --brand-black).
Pick the right level based on the request:
If user asks for “everything”, deliver in this order:
For UI reviews, check these first:
#FFFFFF and dark #050505; don't force the wrong one)When doing strict standards audits, fetch current web interface guidance from:
references/web-interface-guidelines-source.md (contains canonical source URL).
Use these resources on demand (don’t load everything unless needed):
brand.json, tokens.css)docs/aesthetic-tokens.md → full Starchild visual spec (mirrors the registry)docs/CREDITS.md → attribution and lineagereferences/ → UX frameworks, checklists, interaction patternsdata/ → palettes, typography, component/pattern datasets, stack-specific CSVsscripts/design_system.py → quick token/system generationscripts/search.py + scripts/core.py → utility helpers for the bundled datasettools
TQX (tqx.trade) HK/US stock quant workflow via tqx-cli: cross-sectional factor analysis, event-driven strategy backtests on the panda_backtest engine, and agent-driven automated paper trading. Use when the user wants to run factor IC/IR analysis, backtest a Python trading strategy on Hong Kong or US stocks, or set up agent-automated trading (e.g. "backtest a moving-average strategy on AAPL", "analyze a momentum factor on HK stocks", "let the agent trade my paper account").
tools
中学物理实验教学参赛方案套件。包含四个完整实验(单摆测g、声悬浮测声速、向心力定量演示仪、电磁阻尼定量研究),每套含固件源码、3D打印图纸、教学PPT、教学文稿、采购清单、调试手册、视频分镜脚本、模拟器。Use when a teacher/student needs a complete, classroom-ready physics experiment package for competition or teaching.
development
End-to-end blog management for AI agents. Write, import, build, preview, and publish articles using the blog template. Covers draft workflow, article import, SEO, OG images, and deployment.
development
ESP8266/ESP32 + 磁簧开关/霍尔传感器 单摆测重力加速度实验套件。 硬件搭建→固件烧录→数据采集→Web仪表盘→g值计算→教学文档,一站式完成。 Use when the user wants to build a pendulum g-measurement experiment, measure gravity with a micro-controller, or needs a complete physics experiment package with hardware + firmware + dashboard + teaching materials.