Toolsday

October 27, 2026 · Last updated on September 12, 2026

AI Toolsday: Real AI, Real Agile, Real Workflows

# Continuous Improvement

Practical AI workflows for Agile teams

Build Your Own AI Agile Coach

What if you had an Agile thought partner available whenever you needed to work through a problem?

One that knew your role, understood the context you're working in, challenged your assumptions, asked you better questions, and helped you think through your next move?

For our final AI Toolsday workflow of the season, we're putting everything together.

This week, you'll create your own AI Agile Coach—a reusable AI conversation designed to help you think through challenges rather than simply hand you answers.

📌 Before You Start

The AI tools featured in this workflow are examples, not requirements. Every organization has different approved tools, security policies, and ways of working.

The goal of AI Toolsday isn't to recommend one specific product—it's to share practical AI workflows that you can adapt using the tools available to you.

Focus on the workflow, not the tool.

🛠️ Tools Used

ChatGPT or Claude

Use your organization's approved conversational AI tool to create an ongoing coaching conversation that understands how you want it to help you think.

Your Own Context

The most important ingredient isn't another AI tool—it's the context you provide about your role, responsibilities, goals, constraints, and ways of working.

The better the context, the more useful your coach can become.

⚙️ How It Works

Step 1: Define what you want your coach to do

Don't start by telling AI:

"Be my Agile coach."

That's too broad.

Instead, think about where you could actually use a thought partner.

Maybe you want help:

Preparing for difficult conversations

Facilitating team challenges

Thinking through impediments

Challenging assumptions

Improving flow

Preparing for PI Planning

Coaching teams toward greater ownership

Exploring product decisions

Reflecting on your own leadership

Deciding what question to ask next

Choose the areas where better thinking would help you most.

Step 2: Give your coach context about you

Next, explain your role and environment.

You might include:

Your Agile role

Who you work with

Your responsibilities

Your team's or ART's current challenges

Goals you're working toward

Constraints you regularly face

Areas where you're trying to improve

You don't need to provide confidential details. Give AI enough context to understand the environment without sharing information your organization wouldn't want entered into the tool.

Step 3: Tell it HOW to coach you

This might be the most important step.

If you simply ask AI for advice, you'll probably get advice.

Instead, establish some coaching behaviors.

For example:

Don't immediately give me a solution.

Ask questions before making recommendations.

Challenge my assumptions when appropriate.

Point out when I may be treating a symptom instead of the underlying problem.

Give me multiple perspectives.

Tell me when there isn't enough information to reach a conclusion.

Help me decide what I should learn or investigate next.

You're teaching AI that its job isn't to solve everything for you.

Its job is to help you think better.

Step 4: Bring it a real challenge

Now give your coach something you're actually working through.

For example:

"Our team regularly carries work from one Sprint into the next. Help me think through what's happening."

Instead of immediately asking for solutions, let your AI coach question you.

It might ask about work size, interruptions, dependencies, WIP, capacity, refinement, Sprint Goals, or other factors.

Answer the questions.

Then keep going.

The value often comes from the conversation, not the first response.

Step 5: Ask it to challenge you

Once you've explored the problem, deliberately introduce some friction.

Try:

"What assumptions am I making?"

"What would someone on the team disagree with?"

"What evidence would tell us our hypothesis is wrong?"

"Are we solving the right problem?"

"What's another way to interpret what's happening?"

This keeps your AI coach from becoming an agreement machine.

Step 6: End with an experiment

Before you finish the conversation, ask:

"What's one small thing I could try?"

Not a massive transformation plan.

Not twelve recommendations.

One experiment.

Define what you'll try, what you expect to learn, and what you'll look for to determine whether it's helping.

Then take that experiment back into the real world.

💡 Why Agile Pros Care

AI is incredibly good at generating answers.

But Agile professionals don't always need more answers.

Sometimes we need better questions.

A well-designed AI coach can give you a private space to think through a problem, examine assumptions, consider different perspectives, prepare for conversations, and explore possible experiments before deciding what to do.

And unlike a generic prompt you use once, you can return to the same coaching conversation as your situation evolves.

AI shouldn't replace your Scrum Master, coach, mentor, team, or community.

But it can become another thinking partner in your toolkit.

💬 Prompt to Copy

You are my AI Agile Coach.
Your job is not to immediately solve problems for me. Your job is to help me think more clearly about the challenges I'm facing.
When I bring you a challenge:
Ask me questions to understand the context before recommending solutions.
Help me separate facts from assumptions.
Challenge my interpretation when appropriate rather than automatically agreeing with me.
Help me consider the perspectives of the people involved.
Identify possible systemic factors—not just the most obvious symptom.
Tell me when there isn't enough evidence to reach a conclusion.
Help me identify what additional information I should gather.
When appropriate, suggest several possible approaches rather than one "right" answer.
End by helping me identify one small experiment or next step.
Keep your coaching practical and grounded in Agile and Lean ways of working.
About me: My role: [ROLE] The people/teams I support: [CONTEXT] My primary responsibilities: [RESPONSIBILITIES] Current goals: [GOALS] Challenges I'm currently navigating: [CHALLENGES]
Once you've read this context, don't give me advice yet. Ask me:
"What would you like to work through today?"

✅ Try It This Week

Create your AI Agile Coach using the prompt above and bring it one real challenge you're currently facing.

But give yourself one rule:

Don't ask it for the answer.

Let it ask you questions first.

See whether you walk away thinking differently about the problem than you did when you started.

💬 Continue the Conversation

If you had an AI Agile Coach available whenever you needed it, what would you ask it to help you with first?

Share your answer below. You might discover someone else in the community is working through the exact same thing.

🤝 AI Toolsday Tip

The best AI assistant isn't necessarily the one that gives you the fastest answer. Sometimes it's the one that helps you slow down enough to ask a better question.

Use AI to challenge your thinking, explore possibilities, and prepare for action—but keep judgment, empathy, context, and decision-making human.

Comments (0)

Popular