Skip to main content

Documentation Index

Fetch the complete documentation index at: https://mintlify.com/aws-samples/sample-well-architected-skills-and-steering/llms.txt

Use this file to discover all available pages before exploring further.

Both methods on this page require no local clone of the repository. They download only what they need, run the installer in a temporary directory, and clean up afterwards. Both also auto-detect which AI coding tools are configured in your current project — so if you have .cursor/, .claude/, .kiro/, and .junie/ present, all four tools are configured in one command.

skills.sh via npx

skills.sh is a registry and CLI for AI agent skills that follows the open Agent Skills specification. Running npx skills add pulls the skill package directly from the GitHub repository and installs it into your project without any global installation step.
# Auto-detect tools and install all skills
npx skills add aws-samples/sample-well-architected-skills-and-steering

# List available skills before installing
npx skills add aws-samples/sample-well-architected-skills-and-steering --list

# Install a specific skill only
npx skills add aws-samples/sample-well-architected-skills-and-steering --skill wa-review

# Install globally (applies to all projects, not just the current one)
npx skills add aws-samples/sample-well-architected-skills-and-steering -g
Use --list first to preview all available skills and choose only the ones relevant to your project. The wa-review skill is the most comprehensive — it covers all six Well-Architected pillars with 307 best practices.

Bootstrap script

The bootstrap script downloads the repository as an archive, extracts it to a temporary directory, runs install.sh (or install.ps1) with the arguments you pass, and removes the temporary directory on exit. With no arguments it defaults to --tool auto --force, which auto-detects every AI tool configured in the current directory and installs for each one, overwriting any existing files without prompting.
# Auto-detect tools in the current directory
curl -sL https://raw.githubusercontent.com/aws-samples/sample-well-architected-skills-and-steering/main/bootstrap.sh | bash

# Install for a specific tool
curl -sL https://raw.githubusercontent.com/aws-samples/sample-well-architected-skills-and-steering/main/bootstrap.sh | bash -s -- --tool kiro

# Install for multiple tools
curl -sL https://raw.githubusercontent.com/aws-samples/sample-well-architected-skills-and-steering/main/bootstrap.sh | bash -s -- --tool kiro --tool claude-code

# Install for all supported tools
curl -sL https://raw.githubusercontent.com/aws-samples/sample-well-architected-skills-and-steering/main/bootstrap.sh | bash -s -- --tool all

What the bootstrap script does

1

Download the repository archive

The script fetches the repository as a .tar.gz (bash) or .zip (PowerShell) from GitHub and extracts it into a uniquely named temporary directory. No git installation is required.
2

Auto-detect configured AI tools

When no --tool flag is passed, the installer scans the current directory for tool configuration markers — .cursor/, .claude/ or CLAUDE.md, .kiro/, .junie/, .openclaw/, .agents/, .clinerules, .windsurfrules, .github/, GEMINI.md, and AGENTS.md — and installs for every tool it finds.
3

Copy skills and steering files

For each detected tool, the installer copies the appropriate adapter files (steering instructions, rules, slash commands) and all skill SKILL.md files into the tool’s expected directories within your project.
4

Clean up

The temporary directory is removed automatically via a trap EXIT (bash) or finally block (PowerShell), regardless of whether the installation succeeded or failed.
The bootstrap script does not support symlinks. If you want to use --symlink mode so that future updates to the repository automatically propagate to your project without reinstalling, use the local install script instead.

Build docs developers (and LLMs) love