Agent Skills FAQ

Everything you need to know about production-grade engineering skills for AI coding agents

Comprehensive answers to common questions about our development lifecycle, skills, integrations, and implementation

73,371
Stars on GitHub
7,930
Forks
146
Open Issues
24
Production Skills

Frequently Asked Questions

Find detailed answers about Agent Skills, how they work, and how to implement them in your development workflow.

General Overview

What are Agent Skills?

Agent Skills are structured workflows that encode the best practices and engineering judgment of senior developers. They're designed to be used by AI coding agents to ensure production-quality code across all phases of development.

Each skill follows a consistent anatomy with clear processes, verification gates, and anti-rationalization tables to prevent shortcuts that compromise quality.

Why should I use Agent Skills?

AI coding agents often default to the shortest path, skipping critical practices like specifications, testing, and security reviews. Agent Skills enforce the same discipline that senior engineers bring to production code.

They bake in best practices from Google's engineering culture, concepts from Software Engineering at Google, and proven industry standards to ensure reliable, maintainable code.

What makes these skills "production-grade"?

Production-grade means:

  • Specific: Actionable steps rather than vague advice
  • Verifiable: Clear exit criteria with evidence requirements
  • Battle-tested: Based on real-world engineering workflows
  • Minimal: Only what's needed to guide the agent

Skills include concepts like Hyrum's Law in API design, the Beyonce Rule in testing, trunk-based development in git workflows, and feature flags in CI/CD.

Skills & Commands

How many skills are included?

There are 24 production-grade skills total: 23 lifecycle skills plus the meta-skill using-agent-skills. Skills are organized across the development lifecycle:

  • Meta: 1 skill for mapping work to appropriate workflows
  • Define: 3 skills for clarifying requirements
  • Plan: 1 skill for task breakdown
  • Build: 7 skills for implementation
  • Verify: 2 skills for testing and debugging
  • Review: 4 skills for quality assurance
  • Ship: 6 skills for deployment and maintenance
What are the slash commands?

There are 8 slash commands that map to the development lifecycle:

  • /spec - Define what to build (Spec before code)
  • /plan - Plan how to build it (Small, atomic tasks)
  • /build - Build incrementally (One slice at a time)
  • /test - Prove it works (Tests are proof)
  • /review - Review before merge (Improve code health)
  • /webperf - Audit web performance (Measure before optimize)
  • /code-simplify - Simplify the code (Clarity over cleverness)
  • /ship - Ship to production (Faster is safer)

Commands like /build auto can generate and implement tasks autonomously while maintaining verification.

How do skills activate automatically?

Skills activate based on context and activity:

  • Designing an API triggers api-and-interface-design
  • Building UI components triggers frontend-ui-engineering
  • Writing tests activates test-driven-development
  • Creating pull requests engages code-review-and-quality

The meta-skill using-agent-skills maps incoming work to the right skill workflow based on the task context.

Implementation & Setup

How do I install Agent Skills?

The fastest installation method uses the skills CLI:

npx skills add addyosmani/agent-skills # install all 24 skills
npx skills add addyosmani/agent-skills --list # browse before installing

For individual skills:

npx skills add addyosmani/agent-skills --skill code-review-and-quality # five-axis review
npx skills add addyosmani/agent-skills --skill interview-me # requirements interrogation
npx skills add addyosmani/agent-skills --skill test-driven-development # red-green-refactor
Which AI coding agents support Agent Skills?

Agent Skills integrates with 70+ AI coding agents including:

  • Claude Code (recommended)
  • Cursor
  • Antigravity CLI
  • Gemini CLI
  • Windsurf
  • GitHub Copilot
  • OpenCode
  • Kiro IDE & CLI
  • Codex / Other Agents

Each tool has specific setup instructions in the documentation.

What are the system requirements?

Requirements vary by integration:

  • CLI: Node.js 16+ with npm
  • Claude Code: Claude CLI with marketplace access
  • Cursor: Cursor IDE with skill files in .cursor/rules/
  • Antigravity CLI: Antigravity CLI with plugin system
  • Gemini CLI: Gemini CLI with skill support

All skills are written in plain Markdown and work with any agent that accepts system prompts or instruction files.

Development Lifecycle

How does the development lifecycle work?

The lifecycle follows a structured approach:

DEFINE PLAN BUILD VERIFY REVIEW SHIP ┌──────┐ ┌──────┐ ┌──────┐ ┌──────┐ ┌──────┐ ┌──────┐ │ Idea │ ───▶ │ Spec │ ───▶ │ Code │ ───▶ │ Test │ ───▶ │ QA │ ───▶ │ Go │ │Refine│ │ PRD │ │ Impl │ │Debug │ │ Gate │ │ Live │ └──────┘ └──────┘ └──────┘ └──────┘ └──────┘ └──────┘ /spec /plan /build /test /review /ship

Each phase has dedicated skills that enforce best practices and verification gates before proceeding to the next phase.

What happens during the "Define" phase?

The Define phase focuses on clarifying requirements:

  • interview-me: Extracts actual requirements through structured questioning
  • idea-refine: Explores and refines rough concepts
  • spec-driven-development: Creates detailed PRDs before any code

This phase ensures you're building the right thing with clear specifications and acceptance criteria.

How does the "Ship" phase ensure safety?

The Ship phase includes multiple safety mechanisms:

  • ci-cd-and-automation: Feature flags and quality gate pipelines
  • shipping-and-launch: Staged rollouts and rollback procedures
  • observability-and-instrumentation: Structured logging and tracing
  • deprecation-and-migration: Code-as-liability mindset

The principle "Faster is safer" means automated deployments with proper monitoring are safer than manual releases.

Quality & Best Practices

How do skills ensure code quality?

Quality is enforced through multiple mechanisms:

  • Verification gates: Every skill requires evidence (tests passing, build output, runtime data)
  • Anti-rationalization tables: Counterarguments against common excuses for skipping steps
  • Five-axis code review: Comprehensive review covering multiple quality dimensions
  • Test pyramid enforcement: 80/15/5 split between unit/integration/e2e tests

Skills like security-and-hardening and performance-optimization include specific quality checks.

What testing practices are included?

The test-driven-development skill enforces:

  • Red-Green-Refactor cycle
  • Test pyramid (80% unit, 15% integration, 5% e2e)
  • DAMP over DRY (Descriptive And Meaningful over Don't Repeat Yourself)
  • Beyonce Rule (if it's not tested, it doesn't exist)
  • Browser testing with DevTools MCP integration

Tests are treated as proof of functionality, not just documentation.

How is security handled?

The security-and-hardening skill includes:

  • OWASP Top 10 prevention
  • Authentication patterns and secrets management
  • Dependency auditing
  • Three-tier boundary system
  • Input validation and sanitization
  • CORS and security headers

Security is treated as a first-class concern throughout the development lifecycle.

Integrations & Compatibility

How do I integrate with Claude Code?

Two methods for Claude Code:

Marketplace install:

/plugin marketplace add addyosmani/agent-skills
/plugin install agent-skills@addy-agent-skills

Local development:

git clone https://github.com/addyosmani/agent-skills.git
claude --plugin-dir /path/to/agent-skills

SSH errors can be resolved by adding SSH keys or using HTTPS URLs.

What about GitHub Copilot integration?

For GitHub Copilot:

  • Use agent definitions from agents/ as Copilot personas
  • Add skill content to .github/copilot-instructions.md
  • Skills act as enhanced system prompts for better code generation

This creates a hybrid approach where Copilot generates code while following structured workflows.

Can I use skills with other tools?

Yes, skills are designed for broad compatibility:

  • Plain text: All skills are Markdown files that work with any AI tool
  • Custom integrations: Add skill files to your IDE's rules directory
  • API-first: Skills can be integrated via the skills CLI API

See the documentation for integration guides for your specific tool.

Contributing & Community

How can I contribute to Agent Skills?

Contributions are welcome! Follow these guidelines:

  • Skills should be specific, verifiable, battle-tested, and minimal
  • See docs/skill-anatomy.md for the format specification
  • Follow the contribution guidelines in CONTRIBUTING.md
  • Start with small improvements to existing skills
  • For new skills, propose them in issues first

All contributions follow the MIT license and are reviewed by maintainers.

Where can I get help?

Support resources include:

  • GitHub Issues for bug reports and feature requests
  • GitHub Discussions for community questions
  • Comprehensive documentation in the docs/ directory
  • Setup guides for each supported tool

Join our community of 422+ contributors working on improving AI development practices.

What's the roadmap for Agent Skills?

Current focus areas include:

  • Expanding integrations to more AI coding tools
  • Adding more specialized skills for specific domains
  • Improving skill orchestration for multi-agent systems
  • Enhanced performance optimization workflows
  • Better security hardening practices

Roadmap is driven by community feedback and emerging industry needs. Check GitHub Issues for planned features.