Personal AI agent skills for Abelion's development workflow. Based on Anthropic Skills standard.
Modular skills that extend AI agent capabilities with context, conventions, and structured workflows — built for the Abelion stack.
| Skill | Status | Description |
|---|---|---|
| vibe-coding | ✅ Ready | Flow state coding workflow |
| code-review | ✅ Ready | Code quality & best practices |
| security-review | ✅ Ready | OWASP-based security audit |
| performance | 🔜 Planned | Performance optimization |
| design-review | 🔜 Planned | UI/UX & accessibility |
Each skill follows Anthropic's standard structure:
skill-name/
├── SKILL.md ← Step-by-step mode (you stay in control)
├── SKILL.agent.md ← Agentic mode (delegate to agent)
├── README.md ← Skill documentation
└── references/ ← Context files: cheatsheets, patterns, docs
| IDE | Support |
|---|---|
| Antigravity | ✅ Via Skills settings |
| Zed | ✅ Via context_servers |
| Claude.ai | ✅ Via project knowledge |
| Claude Code | ✅ Via /skill command |
| Cursor / Windsurf | ✅ Via MCP or context |
All skills are optimized for:
- Runtime: Bun 1.x
- Frontend: React 19 + Vite 6 + TypeScript 5 + Tailwind CSS 4 + shadcn/ui
- Backend: Supabase (PostgreSQL + Auth + Edge Functions + RLS)
- State: TanStack Query 5 + Zustand 5
- Validation: Zod 3
- Animation: GSAP 3 + Framer Motion 11
- Deploy: Vercel
See CONTRIBUTING.md for how to use and add skills.
Built by Abelion512 · Inspired by anthropics/skills