THE CODE CHANGE REVIEW
Understanding AI HQ free worksheet, version 1.0 (2026-11-12)
Fill it in online, print it or get the PDF: Understanding AI HQ website, Learn > Worksheets

HOW TO USE IT
1. Fill in sections 1 and 2 before you ask for a change.
2. Fill in section 3 after, before you keep it.
3. Keep a working copy each time the tests pass.

================================================================
1. BEFORE YOU ASK
================================================================

THE ONE CHANGE I WANT
>

THE EXACT ERROR (LAST LINE, PLUS FILE AND LINE)
>
>

THE TEST THAT WILL PROVE IT WORKS
>

================================================================
2. NOTHING PRIVATE IN THE PROMPT
================================================================

I'VE CHECKED MY PROMPT CONTAINS NO
[ ] API keys, passwords or tokens
[ ] .env file contents
[ ] Real customer or personal data (use a fake sample)
[ ] Confidential code I'm not allowed to share

================================================================
3. BEFORE YOU KEEP IT
================================================================

WHAT CHANGED, IN ONE SENTENCE
>
>

REVIEW
[ ] The test that failed now passes
[ ] All the other tests still pass
[ ] The change touches only what I asked for
[ ] I can explain every line that changed
[ ] I've saved a working copy

A CASE THE TESTS STILL DON'T COVER
>

----------------------------------------------------------------
Passing tests mean the code does what you tested, not that it has no bugs. Never paste a secret into a prompt; if you do, cancel it and make a new one.


################################################################
# EXAMPLE ANSWERS
################################################################

THE CODE CHANGE REVIEW
Understanding AI HQ free worksheet, version 1.0 (2026-11-12)
Fill it in online, print it or get the PDF: Understanding AI HQ website, Learn > Worksheets

HOW TO USE IT
1. Fill in sections 1 and 2 before you ask for a change.
2. Fill in section 3 after, before you keep it.
3. Keep a working copy each time the tests pass.

EXAMPLE ANSWERS. The example is the real bug and fix from the lesson, run on 12 November 2026 with one Claude model.

================================================================
1. BEFORE YOU ASK
================================================================

THE ONE CHANGE I WANT
> average_rating returns None for an empty list instead of crashing.

THE EXACT ERROR (LAST LINE, PLUS FILE AND LINE)
> ZeroDivisionError: division by zero, ratings.py line 3.

THE TEST THAT WILL PROVE IT WORKS
> test_no_ratings_yet: assertIsNone(average_rating([]))

================================================================
2. NOTHING PRIVATE IN THE PROMPT
================================================================

I'VE CHECKED MY PROMPT CONTAINS NO
[x] API keys, passwords or tokens
[x] .env file contents
[x] Real customer or personal data (use a fake sample)
[x] Confidential code I'm not allowed to share

================================================================
3. BEFORE YOU KEEP IT
================================================================

WHAT CHANGED, IN ONE SENTENCE
> Added a check that returns None when the list is empty, before the division.

REVIEW
[x] The test that failed now passes
[x] All the other tests still pass
[x] The change touches only what I asked for
[x] I can explain every line that changed
[x] I've saved a working copy

A CASE THE TESTS STILL DON'T COVER
> Ratings outside 1 to 5.

----------------------------------------------------------------
Passing tests mean the code does what you tested, not that it has no bugs. Never paste a secret into a prompt; if you do, cancel it and make a new one.