Skillzwave

Quick Start Guide

Get up and running with Skilz in just a few minutes. This guide covers installation and basic usage to help you install your first AI skill.

Flowchart showing Skilz installation process: run install command, registry lookup, git clone, manifest parsing, file installation

Step 1: Install Skilz

First, install Skilz using pip. Skilz requires Python 3.10 or higher.

pip install skilz

Verify the installation:

pip install skilz
skilz --version

Tip: If you prefer isolated installations, use pipx install skilz instead.

Step 2: Browse Skills

Visit Skillzwave.ai to browse the skill marketplace. Each skill page shows the install command you need.

For example, to install the Theme Factory skill:

pip install skilz
skilz install anthropics_skills/theme-factory

Step 3: Install Your First Skill

Let's install a skill. Run this command:

pip install skilz
skilz install anthropics_skills/algorithmic-art

This command:

  1. Resolves the skill from the registry
  2. Downloads the skill files from GitHub
  3. Installs to the default agent (Claude Code)
  4. Creates a manifest to track the installation

Step 4: Verify Installation

List your installed skills:

pip install skilz
skilz list

You should see output like:

Installed Skills (claude)
========================
  algorithmic-art  anthropics_skills/algorithmic-art  v1.0.0  abc1234

Common Commands

Here are the most commonly used Skilz commands:

pip install skilz
# Install a skill (defaults to Claude Code, user-level)
skilz install owner_repo/skill-name

# Install for a specific agent
skilz install owner_repo/skill-name --agent opencode
skilz install owner_repo/skill-name --agent gemini

# Install at project level (in current directory)
skilz install owner_repo/skill-name --project

# List installed skills
skilz list
skilz list --agent opencode
skilz list --project

# Update all skills to latest versions
skilz update

# Remove a skill
skilz remove owner_repo/skill-name

# Read skill content (for agents without native skill loading)
skilz read skill-name

Understanding Skill IDs

Skill IDs follow the format owner_repo/skill-name where underscores separate the GitHub owner and repository name.

Example Skill IDs

  • anthropics_skills/theme-factory — Theme Factory skill from anthropics/skills repo
  • SpillwaveSolutions_architect-agent/SKILL — Architect Agent skill
  • myuser_myrepo/my-skill — Your custom skill

Installing for Different Agents

Skilz supports 14 different AI coding assistants. Use the --agent flag:

pip install skilz
# Install for different agents
skilz install owner_repo/skill-name --agent claude    # Claude Code (default)
skilz install owner_repo/skill-name --agent opencode  # OpenCode
skilz install owner_repo/skill-name --agent codex     # Codex
skilz install owner_repo/skill-name --agent gemini    # Gemini (project-only)
skilz install owner_repo/skill-name --agent copilot   # GitHub Copilot (project-only)
skilz install owner_repo/skill-name --agent cursor    # Cursor (project-only)

See the Supported Agents guide for the complete list and agent-specific instructions.

User vs Project Level

Skills can be installed at two levels:

User Level (Default)

Installed to ~/.claude/skills/ (or equivalent for other agents). Available globally for all your projects. 

skilz install owner_repo/skill

Project Level

Installed to ./.claude/skills/ in the current directory. Useful for project-specific skills or sandboxed environments. 

skilz install owner_repo/skill --project

Next Steps

Now that you have Skilz installed, explore these resources:

Need help? Check the Troubleshooting Guide or open an issue on GitHub.