ChatGPT is all about getting creative. While you can give the chatbot a simple query or prompt, some playing around with your input can go a long way in generating the perfect output.
On the Reddit thread ChatGPTPromptGenius, the user Beginning-Willow-801 has shared what they consider to be 14 "cheat codes" to turn ChatGPT into a powerhouse.
These all revolve around the idea of informing ChatGPT of the workstyle, format, and wordiness that you are looking for. By setting these parameters up front, ChatGPT can work more efficiently.
The user lists a total of 14 different prompts to try out. I tried out all 14, and while I found success with all of them, five stood out above the rest.
Speed-read anything
Prompt:
Workstyle: Fast. Verbosity: Short. Format: Bullets.
Task: Summarize [text or URL].
Rules: 5 bullets, max 12 words each. One “so what” line at the end.
Sometimes, when greeted with pages and pages of dense information, it can feel like a daunting task to get through it.
With this prompt, ChatGPT will quickly formulate a response that explains what is in a block of text. This can be an added file, copied and pasted text, or from a URL.
Of course, if you want a detailed answer, this is not the way to go, but it does work quickly to generate a simple summary. This can then also be followed up by asking for a more expansive summary.
Recall and expand
Prompt:
Workstyle: Thinking High. Verbosity: Long. Format: Report. Task: Read the provided notes/text.
Produce a coherent 2,000-word [report]. Sections: Exec summary → Key insights → Evidence (with inline refs) → Risks → Next steps.
Consistency: unify terminology; resolve contradictions explicitly.
On the opposite end of the spectrum to the above prompt, this one looks to produce detailed and enriched documents after examining text. It will come back with a large 2000-word report, including key insights and evidence, as well as next steps.
This, unsurprisingly, takes a lot longer to produce a response than the above query, but if you want a detailed report on a subject, it's a great way to go about it.
Self-critique and rewrite
Prompt:
Workstyle: Auto. Verbosity: Medium. Format: Before/After.
Task: Draft [X]. Then self-critique and rewrite.
Steps:
A) Draft v1 (<=150 words)
B) Critique: clarity, logic, tone, evidence (bullets)
C) Rewrite v2 (<=120 words), fixes applied
This kind of prompt has been growing in popularity recently. Essentially, it asks ChatGPT (or any of its best alternatives) to re-examine its own work. Instead of just producing an answer, it will create one and then look back over it to refine a better response.
It is very similar to the way that you or I would revise our own work, going back over and making changes to improve the structure and phrasing.
For example, when asked to write an explanation of photosynthesis, ChatGPT produced a simple but effective explanation. It then followed that up with a critique, explaining that it was too textbook-like and over-simplified. Then, a second improved draft followed.
Blend sources into one briefing
Prompt:
Workstyle: Thinking Medium. Verbosity: Balanced. Format: Briefing.
Inputs: [article 1], [notes], [data].
Output:
- What’s true across sources (consensus)
- Where they disagree (and why)
- 5-point POV with implications for [audience]
- One-page action checklist
If you have multiple sources to look through, this prompt can quickly search through each of them, picking out both the similarities and differences.
This can help to create a balanced view of a topic, understanding the connections between different viewpoints. That’s especially useful for school projects or analysis at work, where it's important to see each point of view.
ChatGPT had a tendency to be slightly simplified in its analysis here. If you are unhappy with the answers it produces, asking it to ‘dig deeper’ can produce a better response.
Wear the expert mask
Prompt:
Workstyle: Thinking High. Verbosity: Balanced. Format: Answer + Caveats.
Persona: [named expert or role].
Task: Answer [question] as this expert.
Include:
- Reasoning outline
- Blind spots / what this lens might miss
- Alternative lens: how it would answer differently
This kind of prompt has seen hundreds of variations, reinventing itself over and over again. However, with the inclusion of the workstyle opening, this does produce an effective version.
Give ChatGPT an expert role and a task, and it will try to take on that persona to answer the question. For example, giving it the role of a head programmer to analyse a set of code, or a sub-editor to look through and edit a long document.
I have been using these types of prompts for years and have always found they can force ChatGPT into some impressive answers. Most importantly, it triggers the chatbot to think deeper than it would normally.
Conclusion
These aren’t literal cheat codes, but they are a great way to pull out the full functionality of ChatGPT if you're experimenting with it.
This feels especially true for people using the tool for their work outputs. A lot of these prompts work to examine your text or improve your own work.
If none of these prompts apply to what you need, they can help to inspire other prompts. Take the concept of workstyle and apply it to your own prompts, customising how the AI model changes its behaviour to fit your needs.
Follow Tom's Guide on Google News and add us as a preferred source to get our up-to-date news, analysis, and reviews in your feeds. Make sure to click the Follow button!
More from Tom's Guide
- ChatGPT Vision vs Google Lens — I tested 7 prompts and one was the clear winner
- I use ChatGPT every day — but I can’t stand this one feature
- I use these 5 ChatGPT prompts to future-proof my career in an AI-driven job market — here’s why they work
![HIDevolution [2025] ASUS ROG...](https://images.fie.futurecdn.net/products/60cbb9059616003fe7353df7bc0e56e02dd01b3c-100-80.jpg.webp "HIDevolution [2025] ASUS ROG...")
You must confirm your public display name before commenting
Please logout and then login again, you will then be prompted to enter your display name.
