Quick Start
Let's go from zero to your first working AI agent in under 10 minutes. No code required for steps 1–3.
Pick your starting point
An AI model is the "brain" of your agent. Pick whichever feels most familiar — they all work for the steps that follow.
Claude by Anthropic — excellent at reasoning, long documents, and nuanced writing. Great default choice.
➡️ Go to claude.ai (opens in a new tab) and create a free account.
Free tier available. No credit card required to start.
Give it a real task
Don't just say "tell me about AI." Give it a task you actually do at work. Examples:
- "Here are my last 5 customer emails. Draft a response to each one in a professional tone."
- "Summarize this meeting transcript into 5 bullet points with action items."
- "I have a spreadsheet of invoices. Help me write a formula to flag any that are overdue."
Notice how it completes the work — not just answers a question. That's the shift from chatbot to agent.
Level up: use a tool-enabled agent
Claude and other models can be given tools — the ability to browse the web, run code, read files, or call APIs. This is where agents become genuinely autonomous.
Try Claude.ai Projects (free tier available):
- Create a new project
- Upload a document (a PDF, a spreadsheet, anything)
- Ask it to analyze, extract, or act on the content
Automate a workflow
Ready to go further? Pick one of the Recipes and follow it step by step. Each recipe shows exactly how to set up an agent for a specific business task — no coding required for most of them.
Build your own
If you want to build custom agents and deploy them, head to the Tools section. We'll walk you through the full stack — from picking a model to wiring up your first automated workflow.
The most common beginner mistake: being too vague. An agent is only as good as your instructions. The more specific and contextual your prompt, the better the output. Think of it like briefing a smart contractor — they need to know what you want, why it matters, and what good looks like.
What just happened?
You just experienced the core loop of every AI agent:
Instruction → Model thinks → Model acts → ResultAs you add tools, memory, and automation, that loop becomes more powerful. But it's always that same structure. Everything in this cookbook builds on that foundation.