Free resource · Field guide

Using AI Agent Skills in a Business

A practical field guide to turning one-off agent behaviour into reusable, testable, permissioned capabilities that a business can actually operate.

I explain how I would move from a general-purpose agent with a large prompt to a library of explicit skills: deciding what deserves to become a skill, defining the contract around it, attaching tools and context safely, routing work to the right capability, composing skills without creating chaos, evaluating quality, controlling cost, versioning changes, and building the governance needed for teams to trust agent-driven work.

Start the guide ↗15 chapters · free · no signup required
Who it's for

Built for people making the same decisions I made.

This guide follows the same principles I use when thinking about agentic products: tools should be policy-bound, work should be observable, durable state should be explicit, customer-facing actions may need human approval, and every capability should have a clear operating boundary rather than relying on a model to improvise the entire system from one prompt.

Founders building AI-native products

Engineering teams designing agent architectures

Operations teams automating repeatable business workflows

Agencies building reusable AI delivery systems

Consultants turning expertise into repeatable agent capabilities

Product leaders deciding where agents should and should not act autonomously

Where I can help

AI Agent Skills Architecture Review

I help teams identify which workflows should become reusable agent skills, define the skill and tool boundaries, design routing and approval logic, build an evaluation strategy, and create a skill library that can grow without becoming ungovernable.

Book an agent skills review ↗
Work with me

Ready to work through it?

Start the guide and progress through it at your own pace. Start with a 30-minute call with me.

Book a call with me ↗