Maxx your AI agent skills. One command, zero config.
Scans your project, detects your tech stack, and installs curated AI agent skills automatically. Works with Cursor, Claude Code, Copilot, and any AI coding agent.
- Zero config — one command, no setup, no flags required
- Stack detection — scans
package.json, Gradle files, and config files - Curated registry — only audited, skill files from a maintained registry
- Security-first — SHA-256 verified downloads, no live third-party fetches at install time
- Cross-agent — works with any AI coding agent
npx skillmaxxThat's it. skillmaxx detects your stack, selects the best matching skills, and installs them.
-y, --yes Skip confirmation prompt
--dry-run Show what would be installed without installing
-h, --help Show help message
- Frameworks & UI: React, Next.js, Vue, Nuxt, Svelte, Angular, Astro, Tailwind CSS, shadcn/ui, GSAP, Three.js
- Languages & Runtimes: TypeScript, Node.js, Go, Bun, Deno, Dart
- Backend & APIs: Express, Hono, NestJS, Spring Boot
- Mobile & Desktop: Expo, React Native, Flutter, SwiftUI, Android, Kotlin Multiplatform, Tauri, Electron
- Data & Storage: Supabase, Neon, Prisma, Drizzle ORM, Zod, React Hook Form
- Auth & Billing: Better Auth, Clerk, Stripe
- Testing: Vitest, Playwright
- Cloud & Infrastructure: Vercel, Vercel AI SDK, Cloudflare, Durable Objects, Cloudflare Agents, Cloudflare AI, AWS, Azure, Terraform
- Tooling: Turborepo, Vite, oxlint
- Media & AI: Remotion, ElevenLabs
skillmaxx does not install directly from random upstream repositories at runtime.
Skills are synced by maintainers into the repository-local skill registry, scanned for prompt-injection and supply-chain risks, and recorded with SHA-256 hashes in a manifest. When you run skillmaxx, the CLI downloads only the skills your project needs from that curated registry, verifies every file against the manifest, and writes a skills-lock.json entry with the installed source and bundle hash.
skillmaxx/ → Astro website (docs & marketing)
packages/skillmaxx → the npm CLI package
Contributions are welcome! See CONTRIBUTING.md and our Code of Conduct.
Found a security issue? Read SECURITY.md.