Browse the catalog. Pick what you need. Install with one command. Each skill is independent -- no bundle, no framework, no lock-in.
Works with any AI coding tool that supports agent skills -- Claude Code, Cursor, Windsurf, GitHub Copilot, OpenAI Codex, OpenCode, and more.
Browse the catalog | Install a skill
Pick one skill:
npx skills add https://github.com/luongnv89/skills --skill code-reviewPick several:
npx skills add https://github.com/luongnv89/skills --skill code-review --skill auto-push --skill test-coverageOr grab everything:
npx skills add https://github.com/luongnv89/skillsUse agent-skill-manager (asm) to manage skills across all your AI coding agents from a single TUI/CLI:
npm install -g agent-skill-managerasm install github:luongnv89/skillsasm search # Search skills by name or descriptionasm list # List all installed skillsOther install methods
Remote install (no clone)
Interactive TUI to pick skills, tools, and scope:
curl -sSL https://raw.githubusercontent.com/luongnv89/skills/main/remote-install.sh | bashNon-interactive:
curl -sSL https://raw.githubusercontent.com/luongnv89/skills/main/remote-install.sh | bash -s -- \
--skills "code-review,auto-push" --tools "Claude Code" --scope globalClone and run locally
git clone https://github.com/luongnv89/skills.gitcd skills && bash install.shEvery skill is standalone. Install one, install ten -- they don't depend on each other.
To install any skill, copy the command below and replace <skill-name> with the skill name from the catalog:
npx skills add https://github.com/luongnv89/skills --skill <skill-name>| Category | What it covers |
|---|---|
| Code Quality | Reviews, optimization, testing, usability |
| Shipping | Git push, CI/CD, releases, VS Code publishing |
| Product Planning | Ideas, naming, PRDs, architecture, task breakdown |
| Frontend and Design | UIs, logos, themes, diagrams |
| Documentation | Docs, README, SEO, open source, agent config |
| App Store | ASO, review guideline compliance |
| Food and Cooking | Quick healthy recipes from photos or ingredients |
| Tooling | CLI builder, local LLMs, scripts, skill management |
| Skill | Version | Effort | What it does |
|---|---|---|---|
| code-review | 1.0.1 |  |
Reviews based on Code Smells + The Pragmatic Programmer. Structured reports by severity |
| code-optimizer | 1.2.0 | |
Finds bottlenecks, memory leaks, caching gaps, concurrency issues |
| test-coverage | 1.2.0 |  |
Targets untested branches and edge cases in your existing test suite |
| usability-review | 1.2.0 | |
Usability reviews using Krug's principles with visual scorecards |
| Skill | Version | Effort | What it does |
|---|---|---|---|
| auto-push | 1.0.0 | |
Stage, commit, push with secret and large-file detection |
| devops-pipeline | 1.0.0 | |
Pre-commit hooks + GitHub Actions for quality gates |
| security-setup | 1.2.2 |  |
Local-first security hardening with pre-commit hooks, offline scanners, reports, and gated CI |
| release-manager | 2.4.0 |  |
Version bump, changelog, tags, GitHub release, PyPI/npm publish |
| vscode-extension-publisher | 1.0.0 | |
Publish VS Code extensions to the marketplace with CI setup |
| Skill | Version | Effort | What it does |
|---|---|---|---|
| idea-validator | 1.2.2 | |
Feasibility and market viability feedback before you build |
| brand-name-checker | 1.2.0 | |
Trademark, domain, social, npm, PyPI, Homebrew, apt -- one pass |
| prd-generator | 1.2.2 | |
Structured PRDs from a description or validated idea |
| tad-generator | 1.2.0 | |
Technical architecture docs with data flow diagrams |
| tasks-generator | 1.2.2 | |
Sprint-ready task breakdowns from your PRD |
| Skill | Version | Effort | What it does |
|---|---|---|---|
| frontend-design | 1.2.0 | |
Production-grade UIs with usability-first design |
| logo-designer | 1.2.0 | |
Professional logos with automatic project context detection |
| theme-transformer | 1.0.0 | |
Reskin any UI into cyberpunk, neon, or digital-dark themes |
| excalidraw-generator | 1.2.0 | |
25+ diagram types as Excalidraw JSON |
| drawio-generator | 1.0.1 | |
Draw.io diagrams with multi-page and C4 support |
| openspec-task-loop | 1.0.0 | |
Spec-first, one-task-at-a-time implementation loop |
| website-cloner | 1.1.0 | |
6-phase clone-and-improve orchestrator: URL → improved Vite/React/Tailwind site |
| ↳ website-analyzer | 1.0.1 | |
Phase 1 — 6-dimension website analysis (UI/UX, category, style, perf, security, SEO) as JSON |
| ↳ website-clone-report | 1.0.1 | |
Phase 2 — Stakeholder-friendly clone report from analysis JSON |
| ↳ website-improvement-prd | 1.1.0 | |
Phase 3 — Improvement PRD from analysis + report |
| ↳ website-implementation-plan | 1.1.0 | |
Phase 4 — Sprint-ready tasks.md from the PRD |
| ↳ website-builder | 1.0.1 | |
Phase 5 — Build the improved Vite/React/Tailwind site from tasks |
| ↳ website-clone-final-report | 1.0.1 | |
Phase 6 — Before/after delivery report comparing original to clone |
| Skill | Version | Effort | What it does |
|---|---|---|---|
| docs-generator | 1.2.0 | |
Restructure scattered docs into a coherent hierarchy |
| readme-to-landing-page | 2.0.0 | |
Transform any README into a landing page (PAS, AIDA, StoryBrand) |
| seo-ai-optimizer | 1.0.1 | |
Technical SEO, structured data, and AI bot accessibility |
| oss-ready | 1.1.0 | |
LICENSE, CONTRIBUTING, CODE_OF_CONDUCT, GitHub templates |
| agent-config | 1.1.0 | |
CLAUDE.md and AGENTS.md following best practices |
| Skill | Version | Effort | What it does |
|---|---|---|---|
| aso-marketing | 1.1.0 | |
Full-lifecycle ASO for Apple App Store and Google Play |
| appstore-review-checker | 1.0.0 | |
Pre-submission audit against 150+ Apple review guidelines |
| Skill | Version | Effort | What it does |
|---|---|---|---|
| quick-healthy-recipes | 1.0.0 | |
Three simple, fast, healthy recipes from food photos, ingredient lists, or cravings |
| Skill | Version | Effort | What it does |
|---|---|---|---|
| cli-builder | 1.0.0 | |
Build production CLI tools via 5-step approval-gated workflow |
| ollama-optimizer | 1.0.1 | |
Tune Ollama for max speed based on your GPU/RAM/CPU |
| install-script-generator | 2.0.0 | |
Cross-platform installers with environment detection |
| github-issue-creator | 1.0.0 | |
Issues from screenshots, emails, bug reports -- with PII redaction |
| opencode-runner | 1.2.0 | |
Delegate tasks to opencode with free cloud models |
| context-hub | 1.0.0 | |
Fetch current API/SDK docs before writing integration code |
| skill-creator | 1.1.0 | |
Create, validate, and package your own skills |
| skill-inventory-auditor | 1.0.0 | |
Find and remove duplicate skill installations |
Most skills live directly under skills/<name>/. Multi-skill products that share a single domain group their pieces in a suite folder — an umbrella skill at skills/<umbrella>/ with its phase skills nested at skills/<umbrella>/<phase-skill>/. The umbrella orchestrates the phases; each phase skill stays independently installable.
The current suite is website-cloner (6 phases nested under skills/website-cloner/, indented in the catalog above with ↳). The installers (install.sh, remote-install.sh) discover both top-level and nested skills, so you can install the whole suite or a single phase.
When building your own multi-skill product, mirror this layout: one umbrella + N phase skills under it.
Do I need all the skills? No. Each skill is independent. Install only what you need.
Which AI tools are supported? Any AI coding tool that supports agent skills. Tested with Claude Code, Cursor, Windsurf, GitHub Copilot, OpenAI Codex, and OpenCode. The installer handles file locations and formats automatically.
Can I create my own skills? Yes. Use the skill-creator skill or follow the Contributing Guide.
How is this different from custom prompts? A skill is a structured workflow with references, templates, and quality checks -- version-controlled and shareable. A prompt is a one-off instruction.
Does this affect my runtime code? No. Skills guide your AI agent during development. Nothing to deploy, no runtime dependencies.
npx skills add https://github.com/luongnv89/skills --skill code-reviewView all skills | Contribute | MIT Licensed
Supported Tool Paths (Manual Installation)
| Tool | Global path | Project path |
|---|---|---|
| Claude Code | ~/.claude/skills/<skill>/ |
.claude/skills/<skill>/ |
| Cursor | ~/.agents/skills/<skill>/ + .cursor/rules/<skill>.mdc |
same, relative |
| Windsurf | ~/.agents/skills/<skill>/ + .windsurf/rules/<skill>.md |
same, relative |
| GitHub Copilot | ~/.agents/skills/<skill>/ + .github/instructions/<skill>.instructions.md |
same, relative |
| OpenAI Codex | ~/.agents/skills/<skill>/ + ~/.codex/AGENTS.md |
same, relative |
| OpenCode | ~/.agents/skills/<skill>/ |
same, relative |
Project Structure
.
├── skills/ # Skill source files
│ └── skill-name/
│ ├── SKILL.md # Skill definition
│ ├── scripts/ # Optional scripts
│ ├── references/ # Optional docs
│ └── assets/ # Optional templates
└── .claude/ # Claude-specific config
Creating New Skills
Use the skill-creator skill or create manually:
---
name: my-skill
version: 1.0.0
description: What it does and when to use it
---
# Instructions for the AI agent...See CONTRIBUTING.md for detailed guidelines.
Contributing
Contributions are welcome. Read the Contributing Guide and Code of Conduct.
Security
See SECURITY.md for reporting vulnerabilities.
Acknowledgements
- frontend-design -- inspired by Anthropic's official frontend-design plugin. Independent implementation with a default style guide and usability principles.
- skill-creator -- customized from Anthropic's official skill-creator (Apache 2.0). Added README.md generation step.
Website -- Claude How-To -- Blog