v1.14.0 · 50+ skills · Live on GitHub Pages · MIT

Stop Rebuilding Agent Skills

Your AI coding agent restarts from zero every session. Agent Skills gives you a browsable catalog of standalone, versioned workflows — including this site from docs/. Install only what you need with one command. Your agent follows the same process every time.

Claude Code · Cursor · Copilot · Codex Open source on GitHub Skills install independently

Every session starts from scratch

You rewrite instructions for code review, releases, and planning on every project. Different chats produce different quality.

Prompt drift

You retype the same checklist. Wording shifts between sessions and the agent's output quality drops.

No durable standard

Without a written workflow on disk, the agent cannot apply the same steps consistently across projects.

Tool fragmentation

Each AI tool stores skills in different locations. You maintain duplicate instructions by hand.

Bad agent output compounds

Vague or varying instructions ship inconsistent patterns into code and team habits.

Review gaps reach main

Without a fixed checklist, security and edge cases get skipped when time is short.

Onboarding repeats the work

New contributors and new projects force you to rebuild prompts from memory or stale chat logs.

Token and focus waste

Pasting long instructions each session burns context and still omits the templates and guardrails that matter.

Install a skill. Run the same workflow every time.

Agent Skills is a catalog of standalone workflows. Pick the capabilities you need. Install with one command. Your agent executes the documented process on every project.

Standalone by design

Install only code-review, or only release-manager. Each skill ships as its own directory with SKILL.md, references, and scripts.

One-command install

npx skills add or asm install. The command places files in the correct path for Claude Code, Cursor, or Copilot.

Documented process

Skills include acceptance criteria, guardrails, templates, and references so the agent has the full context on disk.

Why it beats ad-hoc prompts: version-controlled, shareable with your team, and structured so agents load only the references the task needs.

Three steps to a repeatable agent workflow

Step 01

Browse the catalog

Choose from categories: code quality, shipping, product planning, design, documentation, tooling.

Step 02

Install with one command

Run the npx or asm command. Choose global install or project-local scope.

Step 03

Use in any session

Your agent loads the skill by name and follows the exact workflow, references, and rules.

npx skills add https://github.com/luongnv89/skills --skill code-review

Install several at once: add --skill security-setup --skill clean-code or --skill website-cloner. Tagged release: v1.14.0.

Get the Install Commands

What's new in v1.14.0

Catalog growth, a public GitHub Pages home, and suite-aware installers.

GitHub Pages home

Marketing site built from docs/ and deployed on every push to main.

New & expanded skills

security-setup, clean-code, subagent-creator, viral-product-evaluator, tmux-agent-comms, and the website-cloner suite.

Install & release tooling

Suite-aware install.sh / remote-install.sh; release-manager landing-page updates for future releases.

Full release notes on GitHub

Built for developers who ship with agents

Open source and designed to work in the tools you already use.

50+ skills

Code review, security-setup, clean-code, release automation, website-cloner suite, PRDs — each installs on its own.

Multi-tool support

Installers handle paths for Claude Code, Cursor, Windsurf, GitHub Copilot, OpenAI Codex, and OpenCode.

MIT licensed

Free to use, inspect, fork, and contribute. No strings attached.

Catalog v1.14.0 — GitHub Pages home, suite-aware installers, and new skills since v1.13.0.

Pick what you need today

Eight categories — plus multi-phase suites like website-cloner. Install one skill or ten. They stand alone.

Code Quality Shipping Product Planning Frontend & Design Documentation App Store Tooling

code-review

Structured reviews by severity using Code Smells and Pragmatic Programmer patterns.

auto-push

Stage, commit, and push with secret scanning and large-file detection.

release-manager

Version bump, changelog, tags, GitHub release, and package publish in one workflow.

skill-creator

Author, validate, and package your own skills to the same standard.

security-setup

Local pre-commit secrets, dependency scans, and static analysis with gated CI.

clean-code

bbv Clean Code audit with phased plan and optional HTML report.

website-cloner

Six-phase orchestrator to analyze, plan, and rebuild an improved site.

View Full Catalog on GitHub

Common questions

Do I need all the skills?

No. Each skill is independent. Install only what your current project needs.

Which AI tools are supported?

Any agent tool that supports skills. The installer handles file paths for Claude Code, Cursor, Windsurf, Copilot, Codex, and OpenCode automatically.

How is this different from a custom prompt?

A skill is a structured workflow with references, templates, and quality checks — version-controlled and shareable. A prompt is a one-off instruction that lives in chat history.

Does this affect my runtime code?

No. Skills guide your agent during development. Nothing to deploy, no runtime dependencies in your app.

Can I create my own skills?

Yes. Use the skill-creator skill or follow the contributing guide. Same format, same validation tooling.

Is it free?

Yes. MIT licensed. Install, fork, and contribute at no cost.

Install your first skill in under a minute

Catalog v1.14.0 — no bundle, no framework. One command puts a complete workflow on your machine.

Install code-review

MIT licensed · open source · global or per-project scope