Skip to content
New this month 24 fresh C++, C#, F#, JavaScript, TypeScript and Lua courses just landed. Browse new releases Use code WELCOME10 for 10% off your first order · 14-day refund
WisdomCharms — home
Cart 0 items in cart
Prompt Packs & Templates Intermediate Save 44%

Debugging Prompts That Find the Real Bug

Ninety debugging prompts that push a model past the obvious answer toward the real cause.

Rated 5.0 out of 5 from 2 reviews 42 students

LM Created by Laura Mbeki

  • Last updated August 2026
  • English
  • 55m of material
  • 7 lessons

What you will learn

7 concrete outcomes

Every bullet below is something you will have built, shipped or be able to explain by the time you finish the last lesson.

  • Force competing hypotheses instead of accepting the first confident answer
  • Make the model separate observed evidence from its own assumptions
  • Ask for the cheapest experiment that would eliminate a hypothesis
  • Debug flaky tests and race conditions with prompts built for timing problems
  • Investigate memory growth and leak-shaped symptoms methodically
  • Recognise when the model has run out of evidence and needs more from you
  • Write a short defect note afterwards that your team can actually search

Course curriculum

5 modules · 7 lessons · 55m of material

1 lesson running 6m in total. Each lesson ships with the finished source files and a short written recap, so you can follow along in your own editor and skim the module again later.

Lesson-by-lesson titles, code downloads and exercises live inside the course library you get access to straight after checkout.

2 lessons running 14m in total. Each lesson ships with the finished source files and a short written recap, so you can follow along in your own editor and skim the module again later.

Lesson-by-lesson titles, code downloads and exercises live inside the course library you get access to straight after checkout.

2 lessons running 16m in total. Each lesson ships with the finished source files and a short written recap, so you can follow along in your own editor and skim the module again later.

Lesson-by-lesson titles, code downloads and exercises live inside the course library you get access to straight after checkout.

1 lesson running 11m in total. Each lesson ships with the finished source files and a short written recap, so you can follow along in your own editor and skim the module again later.

Lesson-by-lesson titles, code downloads and exercises live inside the course library you get access to straight after checkout.

1 lesson running 8m in total. Each lesson ships with the finished source files and a short written recap, so you can follow along in your own editor and skim the module again later.

Lesson-by-lesson titles, code downloads and exercises live inside the course library you get access to straight after checkout.

5 modules · 7 lessons

55m total length

Requirements

Short list, and deliberately so. If you meet these you can start today.

  • You can read logs and stack traces from your own stack
  • Access to a chat model such as Claude or ChatGPT

About this course

Ask a model why your code is broken and it will confidently name the first plausible cause. Sometimes it is right. Often it is describing a bug that would exist in a similar program, which is worse than no answer because it feels like progress.

These ninety prompts are built to fight that. They force the model to separate what it can see from what it is assuming, to produce competing hypotheses instead of one, to state the cheapest experiment that would rule each hypothesis out, and to say explicitly when the evidence you pasted is not enough. There are dedicated sets for race conditions, memory growth, flaky tests, misbehaving retries, timezone and encoding faults, and the special misery of a bug that only appears in production.

Seven recorded lessons run three real investigations end to end, including one where the model is wrong twice before the prompts steer it right. You see the whole transcript, not a cleaned-up highlight reel.

The pack ships as Markdown, Notion and plain text, with a printable one-page triage flow. Your access link arrives by email as soon as checkout completes and covers every future update.

Topics covered

Frequently asked questions

Still unsure about something? Write to misteryjj100@gmail.com and a human answers, usually the same working day.

No. The prompts are about reasoning under uncertainty, so they work with any stack. The worked investigations use Python and JavaScript because their traces are widely readable.

No, and the pack argues against it. Several prompts exist purely to help you decide the smallest slice of code and logs worth pasting.

By email, as an access link to the WisdomCharms library, immediately after payment. The link does not expire and includes later revisions.

Checkout is handled on our provider's secure payment page. The moment your payment clears we email your personal access link and access code to the address you used at checkout, and the same link appears in your account library. There is nothing to install and nothing to wait for.

Email misteryjj100@gmail.com within 14 days of your purchase, quote your order number, and we refund the full amount to your original payment method. No form to fill in and no questions about how much of the course you watched.

What students say

Reviews are written by people who bought this course. We publish the critical ones too.

5.0

Rated 5.0 out of 5

Course rating · 2 reviews

Rating distribution

  • 5 stars 100%
  • 4 stars 0%
  • 3 stars 0%
  • 2 stars 0%
  • 1 star 0%
  • MW

    Mateusz Wróbel

    Backend engineer

    Apr 2026
    Rated 5.0 out of 5

    The unedited investigations sold me

    Almost all prompt material shows you the tidy version where it works on the first attempt. Publishing three complete investigations with the dead ends left in is braver and far more useful, because you find out how many turns this really takes.
  • DI

    Deepa Iyer

    Test engineer

    Nov 2025
    Rated 5.0 out of 5

    elimination experiments, not guesses

    the whole idea is to make the model commit to something falsifiable and then go and falsify it yourself. obvious in hindsight. I spent a year asking what is wrong with this and getting confident nonsense back.

Your instructor

LM

Laura Mbeki

Prompt engineer and AI workflow designer

  • 506 students taught
  • 16 courses published
  • 4.3 instructor rating
  • Prompt engineering
  • Prompt libraries
  • AI workflows
  • Prompt evaluation

Laura went independent after seven years of agency work and now designs the prompt libraries that sit behind other people's products. She treats prompting as engineering: versioned prompts, a held-out evaluation set, a regression run before anything ships, and a token budget you have to hit. Her packs are the ones she uses with her own clients — briefing, rewriting, summarising, review — rather than sanitised examples, and each comes with notes on where it fails. She keeps every pack working across ChatGPT, Claude and a small open model, so the technique outlives the model.

See every course by Laura Mbeki
$5 USD $9 Discounted from $9. You save 44 percent.

One-time payment · lifetime access

The WisdomCharms dispatch

One useful email a week. No fluff, no spam.

New course releases, discount codes before anyone else, and a short, practical breakdown of one technique — a prompt pattern, a C++ idiom, a TypeScript trick — that you can use the same day.

  • Subscriber-only launch pricing
  • Unsubscribe in one click
  • We never sell your address

By subscribing you agree to our Privacy Policy. Questions? Write to misteryjj100@gmail.com.