AI Workflows: repeatable systems for yourself or a team

01

From a good prompt to a workflow

In simple words

A recipe card is better than remembering how you cooked something once. A workflow is a recipe card for work: what goes in, what you do in order, and what must come out. Anyone can follow the card, even on a busy day.

Key facts

A workflow here means a written, repeatable set of steps where AI does some of the steps and a person does the others. It is not a robot that runs on its own.
Anthropic's engineering article on building effective agents separates two ideas. Workflows are systems where AI models and tools are orchestrated through predefined code paths. Agents are systems where the AI dynamically directs its own process and tool use. The same article advises finding the simplest solution possible and adding complexity only when needed.
In this course you follow that advice by hand: the steps are fixed and written down, and a person runs them in a chat app. No coding is needed.
Every workflow has six parts:

Part	Question	Example: "Weekly customer-email summary"
Trigger	What starts it?	Every Friday at 3 pm
Input	What goes in, and in what shape?	The week's emails, with names removed
Steps	What happens, in order, and who does each (AI or person)?	Sort, summarise, draft reply, person checks
Output	What must come out?	A one-page summary and 5 draft replies
Review	Who checks, and against what list?	The owner, with the quality gate from course 1.1.2
Owner	Who is responsible if it goes wrong?	The team lead

Good candidates are tasks that repeat, follow nearly the same steps each time, and have an output you can check. One-off or very creative tasks are usually better done as a normal chat.
Poor candidates are tasks where a mistake would be costly and cannot be checked, or where the input contains private data your employer's policy does not allow in AI tools.

Step by step

Look at your task list and time log from course 1.1.2. Pick one task you did at least three times in the last month.
Write the six parts (trigger, input, steps, output, review, owner) as one short line each.
In the steps, mark each one "AI" or "me". Anything involving a decision, a promise or private data should be "me".
Cross out any step you cannot explain in one sentence. Split it or remove it.
Check the input: remove or hide private data the task does not need.
Read the six lines out loud to a friend. If they ask "who checks it?", fix the review line.
Check: someone who does not know your job could tell what starts the workflow, what goes in, and what a good result looks like.

Watch on YouTube

Building more effective AI agents (workflows and agents)Anthropic · YouTube · youtube.com

Videos by other creators on YouTube, for reference. They are not made by or affiliated with The Future Market.

LESSON 01From a good prompt to a workflow

LESSON 01 / ANIMATED LESSON / AI VOICE00:00 / 01:00

Voice

Read the video script

[Host] "A good prompt helps once. A workflow helps every week."
[Screen: six boxes labelled trigger, input, steps, output, review, owner] "Six boxes: what starts it, what goes in, the steps, what comes out, who checks it, and who owns it."
[Screen: steps marked AI or me] "I mark each step: does the AI do this, or do I? Decisions and private data stay with me."
[Text on screen: Simple first] "Start with the simplest version that works. Add more only when you need it."

Try it now

Choose one repeating task and fill in the six parts on one page. Mark every step AI or me.

ClaudeStuck? Copy a ready-made prompt for Claude

Copy this, open Claude, paste it, and Claude will walk you through the lesson step by step.

For a grown-up: Claude is for adults (18+). Copy this, paste it into Claude, and use the answer to explain the lesson to your child.

I'm learning "AI Workflows: repeatable systems for yourself or a team" on The Future Market. Lesson 1: "From a good prompt to a workflow".

What the lesson teaches:
- A workflow here means a written, repeatable set of steps where AI does some of the steps and a person does the others. It is not a robot that runs on its own.
- Anthropic's engineering article on building effective agents separates two ideas. Workflows are systems where AI models and tools are orchestrated through predefined code paths. Agents are systems where the AI dynamically directs its own process and tool use. The same article advises finding the simplest solution possible and adding complexity only when needed.
- In this course you follow that advice by hand: the steps are fixed and written down, and a person runs them in a chat app. No coding is needed.
- Every workflow has six parts:
- Part - Question - Example: "Weekly customer-email summary"
- Trigger - What starts it? - Every Friday at 3 pm
- Input - What goes in, and in what shape? - The week's emails, with names removed

The exercise I'm working on:
Choose one repeating task and fill in the six parts on one page. Mark every step AI or me.

I'm stuck. Please:
1. Explain the main idea in plain, simple words with one everyday example.
2. Walk me through the exercise one small step at a time. Ask me what I've done so far before giving the next step.
3. Don't just give me the final answer. Help me do it myself, and check my work when I share it.

I'm a parent helping my child (about 8 years old) with a lesson called "From a good prompt to a workflow" from the course "AI Workflows: repeatable systems for yourself or a team".

Here's what the lesson says:
A recipe card is better than remembering how you cooked something once. A workflow is a recipe card for work: what goes in, what you do in order, and what must come out. Anyone can follow the card, even on a busy day.

The activity:
Choose one repeating task and fill in the six parts on one page. Mark every step AI or me.

Please explain this so an 8-year-old understands: short sentences, a fun everyday example, and no hard words.
Then give me 3 simple questions I can ask my child to check they understood, and a way to do the activity together, one tiny step at a time.

Open Claude

Sources

Anthropic, Building effective agents: www.anthropic.com/engineering/building-effective-agents

My notes

I learned this

02