The 300-Prompt Pack for Code Review and Refactoring
by Laura Mbeki
Sixty copy-paste prompts for the small coding tasks that fill an ordinary working day.
LM Created by Laura Mbeki
Every bullet below is something you will have built, shipped or be able to explain by the time you finish the last lesson.
4 modules · 4 lessons · 30m 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.
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.
1 lesson running 9m 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 7m 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.
4 modules · 4 lessons
30m total length
Short list, and deliberately so. If you meet these you can start today.
Sixty prompts, written and tested against Claude and ChatGPT, aimed squarely at the unglamorous work that fills a developer day: explaining a function you inherited, renaming a badly named module, turning a stack trace into a short list of hypotheses, and writing the commit message you keep putting off.
What you download is a prompt library in three formats: a single annotated Markdown file, a Notion-ready import you can drop straight into a workspace, and a plain text version for tools that only accept pasted input. Each prompt carries a one-line note on what it is for, which placeholders to replace, and the follow-up question that usually gets you the rest of the way.
Four short walkthrough lessons show the prompts working on a real repository. You will see how to attach a diff so the model stops guessing at context it cannot see, how to make it say plainly when it does not know, and how to stop it inventing library functions that were never in the API.
Delivery is instant. After checkout you receive an access link by email that opens the pack in the WisdomCharms library, and every future revision lands behind that same link.
Still unsure about something? Write to misteryjj100@gmail.com and a human answers, usually the same working day.
Reviews are written by people who bought this course. We publish the critical ones too.
3.9
Rated 3.9 out of 5Course rating · 7 reviews
Ravi Sundaram
Site reliability engineer
the trace-to-hypothesis prompts are the ones I kept. rather than pasting a trace and hoping, you paste it alongside what you already know is true, and what comes back is testable. it has changed how I open a bug ticket.
Karolina Nowak
Backend engineer
I was sceptical that sixty prompts could be worth anything at all, then I used the handover one on a Friday afternoon and got back something my cover could actually follow on the Monday. Sorting the pack by task rather than by which assistant you use is the right decision.
Hugo Marchand
Freelance developer
Half of these I would have arrived at myself eventually. The other half I would not, particularly the framing that asks for an explanation aimed at whoever has to maintain the file next. It removes the blank-box moment, which is most of the battle.
Anneke de Vries
Tech lead
For a month this was excellent. After that I had rewritten most of them against our own codebase and stopped opening the pack at all. Arguably that is the intent, but go in knowing you are buying a starting position rather than a reference you keep for years.
Tobias Grün
The renaming and explaining sets are sharp. The review ones sit well below them in quality; asking a model to find problems in a diff is not a prompt so much as a hope. Everything else justified the pound easily.
Freya Sandberg
Staff engineer
If you have used an assistant daily for six months you already have most of these habits and the pack will not add to them. There is nothing on what to do when a prompt fails and you need to diagnose why, which is the part I would genuinely have paid for. Not a bad product, just not one I needed.
Callum Reid
Software engineer
A lot of what is in here is one instruction with the leading verb swapped, and once you have noticed it you cannot stop noticing it. Whole sections amount to three phrasings of describe this code. I accept that it costs almost nothing, but padding a pack out to hit a round number in the title is still padding, and the index is arranged to make the coverage look wider than it is. I would rather have paid the same for twenty prompts that were actually distinct.
Prompt engineer and AI workflow designer
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.
One-time payment · lifetime access