I used to get so frustrated with ChatGPT. I’d ask for something specific and get back a generic, useless wall of text. It felt like a waste of time.
Then I saw this awesome post from an industry pro that totally changed how I think about prompting. The mind behind it pointed out a simple truth: bad prompts lead to bad results. It’s not the AI, it’s the instructions!
The expert shared a brilliant list of prompting frameworks. They’re like cheat codes for getting exactly what you want from the AI. I was blown away by how simple yet powerful they are!
Here are a few of the frameworks the original poster shared:
- ✅ TREF: Task, Requirements, Expectation, Format. Tell the AI what to do, what it needs, what you want, and how it should look.
- ✅ ROSES: Role, Objective, Scenario, Expected Solution, Steps. Have the AI adopt a role and follow a clear path to the solution.
- ✅ PECRA: Purpose, Expectation, Context, Request, Action. Give the AI the ‘why’ behind your request for a more targeted result.
- ✅ GRADE: Goal, Request, Action, Details, Examples. Clearly state your goal and provide examples to guide the AI.
I think this is a total game-changer. Instead of just winging it, these frameworks give your requests a solid structure. You’re telling the AI not just what to do, but how and why.
This is just a quick look at what the contributor shared. The original post has even more frameworks and an infographic, so be sure to check it out for all the details!