LESSON 05 OF 05 · 90 MIN

Free Capstone: Production Lead Intelligence Agent

Build one small working system that uses all four ideas at once — then break it on purpose, four ways.

You have seen four ideas on their own: how a real system is shaped, asking for a filled-in card, checking parcels at the door, and handing over keys carefully. Now you build one small thing that needs all four at once — which is the only way to find out whether they actually stuck.

So what is this really about?

You are going to build something that receives an enquiry, works out what it is about, and does one useful thing with it. Deliberately small. The interesting part is not the happy path — it is that you will break it on purpose, four specific ways, and it has to behave sensibly each time.

An enquiry is checked, stored, understood, routed and acted on, with each stage handling its own failure.

QUICK CHECK

Which of these four would you expect to be hardest to get right?

Answer: option 2. The happy path works almost immediately. Duplicates, bad input and half-finished work are where the real engineering is — which is why the capstone is marked on those.

The words you will hear

Capstone: The build at the end that uses everything, rather than one more exercise.
Extraction: Pulling named facts out of messy human text: who, what, how urgent.
Routing: Deciding where something goes next, based on what you extracted.
Confidence: How sure the AI claims to be — useful only if you decided in advance what to do when it is low.

What you will be able to do

Build a validated, idempotent webhook endpoint.
Extract structured data from an unstructured enquiry.
Execute one controlled tool with an explicit permission boundary.
Demonstrate correct behaviour under invalid input and tool failure.

Going deeper

Everything at once, deliberately small

Four ideas so far: how a real system is shaped, asking for a form instead of a paragraph, checking parcels at the door, and handing over keys carefully. Now you build one thing that needs all four — which is the only way to find out whether they stuck.

An enquiry arrives, is checked and stored, understood, then acted on once.

The requirements

1A door with checks. POST only. Verify a shared secret, cap the size, validate the shape, and recognise a repeat by the sender’s id.
2Save before thinking. The enquiry is stored before any model call. A later failure must not lose it.
3Ask for a form. Contact details, an intent from a fixed list, urgency, and a confidence value.
4Route by rule. A plain rule over those fields. Low confidence goes to a human — the model does not get to decide how much to trust itself.
5One tool, one boundary. A single action whose limits you can state in a sentence.
6Leave a trail. For any enquiry, you can reconstruct what arrived, what was extracted, what was decided and what ran.

Break it on purpose, four ways

Aspect	You do this	It must do this
Send the same enquiry twice	—	One stored enquiry, one action
Send a payload missing a field	—	Rejected with a clear error, nothing stored
Make the model return a bad shape	—	Retried once, then fails without a half-written record
Make the tool fail	—	Enquiry still stored and flagged, never silently dropped

YOUR TURN

Your tool fails after the enquiry was stored and understood. What is the correct outcome?

A system is finished when it fails properly, not when it works once.

Resources

Starter repository specification — Structure, setup and the four failure tests