ai-agents-skills

v1.6.3 - A modular CLI for distributing reusable AI agent skills across multiple coding assistants.

Install 68 curated skills for React, TypeScript, testing, architecture, and more — to 11 AI agents across two tiers. Features project presets, interactive setup, dependency resolution, version tracking, and seamless multi-model sync.

Browse the full skill catalog at joabgonzalez.github.io/ai-agents-skills

Quick Start

# Interactive mode (recommended)
npx ai-agents-skills add

# Install a preset
npx ai-agents-skills add --preset astro-template

# Install specific skills
npx ai-agents-skills add --skill react --skill typescript

Supported Models

11 agents across two tiers. Install once — all agents stay in sync automatically.

Universal — read from .agents/skills/ natively, no extra setup:

Amp · Cline · Codex · Cursor · Gemini CLI · GitHub Copilot · Kimi Code CLI · OpenCode

Dedicated — also get a symlink in their own directory:

Model	Directory	--model ID
Claude Code	.claude/skills/	claude
Antigravity	.agent/skills/	antigravity
OpenClaw	skills/	openclaw

Commands

add — Install skills

# Interactive (prompts for skills, models, and presets)
npx ai-agents-skills add

# Install a preset
npx ai-agents-skills add --preset astro-template

# Target specific models
npx ai-agents-skills add --model claude --model copilot

# Install specific skills
npx ai-agents-skills add --skill react --skill typescript

# Preview without changes — shows exact paths that would be created
npx ai-agents-skills add --skill react --dry-run

Options:

Flag	Description
-p, --preset <id>	Install a project starter preset
-s, --skill <name>	Install a specific skill (repeatable)
-m, --model <name>	Target a specific model (repeatable)
-d, --dry-run	Preview changes — shows exact symlink paths

list — Show installed skills

npx ai-agents-skills list

sync — Add models or update skills

# Interactive (prompts for actions: add models and/or update skills)
npx ai-agents-skills sync

# Update ALL installed skills to latest versions (no prompts)
npx ai-agents-skills sync --update

# Update specific skills to latest versions
npx ai-agents-skills sync --skill react --skill typescript

# Add models to existing installation
npx ai-agents-skills sync --model copilot --model cursor

# Preview without changes
npx ai-agents-skills sync --dry-run

Options:

Flag	Description
-u, --update	Update all installed skills to latest (no prompts)
-s, --skill <name>	Update a specific skill (repeatable)
-m, --model <name>	Add a specific model (repeatable)
-d, --dry-run	Preview changes without applying

remove — Remove skills with dependency checking

# Interactive — first choose: select skills or remove all (purge)
npx ai-agents-skills remove

# Remove specific skills (with dependency check)
npx ai-agents-skills remove --skill react --skill typescript

# Remove all skill entries from all models + ask about AGENTS.md
npx ai-agents-skills remove --purge

# Skip confirmation prompt (useful for automation)
npx ai-agents-skills remove --skill react --confirm

# Dry-run: same interactive flow, no changes applied
npx ai-agents-skills remove --skill react --dry-run

Options:

Flag	Description
-s, --skill <name>	Remove a specific skill (repeatable)
-m, --model <name>	Target a specific model (repeatable)
-p, --purge	Remove all skill entries from all models; asks separately about AGENTS.md
--confirm	Skip confirmation prompt (works in both normal and dry-run mode)
-d, --dry-run	Same flow as normal — prompts included — but no changes applied

How It Works

npx ai-agents-skills add --skill react

• Clones the skill repository to ~/.cache/ai-agents-skills/
• Resolves dependencies: react → javascript, typescript, code-conventions
• Copies skills to .agents/skills/ — the shared directory all 8 universal agents read natively
• Creates symlinks in dedicated model directories (.claude/skills/, .agent/skills/, skills/)
• Updates AGENTS.md with complete "How to Use Skills" workflow (push context)

Installed Structure

your-project/
├── AGENTS.md                 # Push context — complete workflow for ALL models ✨
├── .agents/skills/           # Shared directory — all 8 universal agents read here natively
│   ├── react/
│   ├── typescript/
│   └── code-conventions/
├── .claude/skills/           # Symlinks → .agents/skills/* (Claude Code)
└── .agent/skills/            # Symlinks → .agents/skills/* (Antigravity)

Available Skills (68)

Frameworks

React, Next.js, Astro, Express, Nest, Hono, React Native, Expo

Testing

Jest, Playwright, React Testing Library, React Native Testing Library, E2E Testing, Unit Testing, Testing Strategy

Standards

TypeScript, JavaScript, HTML, CSS, TailwindCSS, A11y

Web

Web Performance, Web SEO

Backend

Node.js, Express, Nest, Hono, Bun, Backend Development

Build Tools

Vite, Webpack

Libraries

MUI, AG Grid, Redux Toolkit, Stagehand

Quality & Architecture

Code Conventions, Code Quality, Form Validation, Critical Partner, Code Refactoring

Architecture Patterns

Architecture Patterns, SOLID, Clean Architecture, Domain-Driven Design, Hexagonal Architecture, Screaming Architecture, Result Pattern, DRY Principle, Mediator Pattern, Sidecar Pattern, Composition Pattern, Circuit Breaker Pattern, State Machines Pattern

Behavioral

English Writing, Technical Communication, Humanizer, Frontend Development, Brainstorming, Systematic Debugging, Interface Design, Writing Plans, Code Review, Verification Protocol, Plan Execution, Subagent Orchestration, Slides

Meta (creation tools)

Skill Creation, Agent Creation, Reference Creation, Prompt Creation, Skill Sync

Creating Skills

Each skill is a directory with a SKILL.md:

skills/my-skill/
├── SKILL.md          # Required — frontmatter + content
└── references/       # Optional — detailed guides

---
name: my-skill
description: "Short description. Trigger: When to activate this skill."
license: "Apache 2.0"
metadata:
  version: "1.0"
  skills:
    - code-conventions
---

# My Skill

## When to Use

...

## Critical Patterns

...

## Decision Tree

...

Telemetry

ai-agents-skills collects anonymous usage data to help prioritize new skills and improvements. No personally identifiable information is ever collected.

To opt out:

ai-agents-skills telemetry disable  # disable
ai-agents-skills telemetry enable   # re-enable
ai-agents-skills telemetry status   # check current setting

What IS collected:

• Names of skills installed
• Preset name used (if any)
• Number of target models
• Whether --dry-run was used
• CLI version, Node.js version, OS platform

What is NEVER collected: file paths, usernames, hostnames, environment variables, or any personally identifiable information.

Telemetry is automatically disabled in CI environments (CI, GITHUB_ACTIONS, CONTINUOUS_INTEGRATION). Data is sent to a private PostHog instance and never shared with third parties.

Troubleshooting

Stale cache after an update or clone issue:

rm -rf ~/.cache/ai-agents-skills

This removes the local skill cache. The next npx ai-agents-skills run re-downloads it fresh.

License

Apache 2.0