Build an AI Agent · Lesson 1Free~12 min readNo code requiredOne task delegated by the end

From chatbot to agent — your first task agent.

Most people use AI like a vending machine: ask, get an answer, paste it somewhere. An agent works differently — you hand it a goal and it does the steps for you. This lesson shows what actually makes something an agent, which tasks to delegate first, and how to set one up for a real job.

The mental model

A chatbot answers. An agent acts. The shift is from “ask and copy” to “delegate and review.”

When you chat with AI, you ask a question and paste the answer somewhere yourself. An agent flips that: you give it a goal, it makes a plan, takes the steps, and hands you a finished result. You stop being the one doing each step and become the one checking the work.

The Reframe

You’re a manager now, not a typist. A good agent task is one you’d hand to a capable assistant with clear instructions. If you can write the instructions, you can build the agent.

Step 01 What actually counts as an agent

Four things separate an agent from a chat:

Business examples: “research these 20 companies and give me a one-line summary of each,” “go through this folder and pull every invoice into a table,” “check this inbox each morning and tell me what needs a reply.”

Step 02 Set up your first task agent

Pick one repetitive, multi-step task that’s low-risk to get slightly wrong. Then write it like a brief for an assistant:

Your agent brief

  1. Goal — the end result you want, stated plainly.
  2. Steps — the rough sequence you’d follow yourself.
  3. Done looks like — exactly what a finished result should be.
  4. Constraints — what it must not do, and when to stop and ask you.
Agent setup promptAct as an agent for this task. Goal: [goal]. Work through it step by step: [steps]. A finished result looks like [done]. Constraints: [what not to do]. Before doing anything that can’t be undone, stop and check with me. Show your plan first, then carry it out, then summarize what you did.

Step 03 Make the first one easy to trust

Your first agent should be narrow and safe. Good first tasks are:

Left vague, agents wander — they over-research, invent steps, or take an action you didn’t want. Keep the first one narrow and require it to confirm before anything irreversible. Autonomy is earned, not granted on day one.

Your challenge: delegate one real task

Pick a repetitive, multi-step task you normally do by hand. Then:

  1. Write the agent brief (goal, steps, done, constraints).
  2. Have the agent show its plan before it acts.
  3. Let it run, then review the finished result.
  4. Note where it wandered — that’s what you’ll tighten next time.

That’s one real task off your plate, done by an agent you directed. Next, give it tools — your apps and data — so it can do far more than talk. That’s Lesson 2.

What you can do now

  • Explain the difference between a chatbot and an agent
  • Recognize the four traits of a real agent task
  • Write an agent brief: goal, steps, done, constraints
  • Set up an agent that plans, acts, and reports back
  • Pick safe, narrow first tasks and keep a human in the loop
Pro
Up next in Build an AI Agent

Lesson 2 · Give your agent tools — connect apps & multi-step work

An agent is only as capable as the tools it can reach. Connect it to your apps and data — safely — and chain real multi-step workflows. Go to Lesson 2 →

🎓
AI Coach
Ask anything about this lesson
Hey! I’m your AI Coach for this lesson. Ask me anything about agents — what counts as one, which tasks to delegate first, or how to write a good agent brief. What’s on your mind?
Free lesson coaching is limited to 3 questions. Upgrade to Pro for unlimited coaching on every lesson.