PromptItIn logoPromptItIn
Log in
Free guide

Learn Prompt Engineering
in 5 minutes.

What a prompt is, why most of them fail, and the 5-part framework that fixes it — with live examples on your own prompts.

Start here

What is a prompt?

A prompt is the message you send to an AI. When you open ChatGPT and type something, that text is your prompt. The AI reads it, processes it, and responds. Everything about that response — the quality, the tone, the relevance, the format — is shaped entirely by what you wrote.

Most people treat it like a Google search: a few keywords, a quick question. That works for search engines because search has been trained to guess what you mean. AI hasn't been trained to guess what you need. It takes your prompt literally.

Same AI. Same question. See the difference.

✕

Typical prompt

help me with my presentation

The AI will produce something generic. You'll rewrite most of it.

✓

Structured prompt

I'm a product manager presenting a new feature to our engineering team on Friday. Write a 3-minute verbal opening that covers the business context, why this matters now, and what we're asking the team to build. Conversational tone — no slides, just talking points.

The AI has everything it needs. The output is usable, not just readable.

Why it matters

What is prompt engineering?

Prompt engineering is the practice of writing prompts intentionally — not just typing what comes to mind, but structuring your request in a way the AI can act on precisely.

It sounds technical. It isn't. You don't need to know anything about AI models, training data, or machine learning. You just need to give the AI what it needs: who it's talking to, what exactly you want done, what rules to follow, and how to format the answer.

The AI is literal.

It responds to what you wrote, not what you meant. Vague in, vague out. The more specific you are, the more useful the response.

Same AI, different results.

Two people asking the same AI the same question get completely different answers — because their prompts are different. The model is identical. The prompt isn't.

You don't need a course.

Prompt engineering is just structured thinking. Once you know the 5 elements, you'll do it naturally — in under a minute.

The good news: there are exactly 5 things that make a prompt go from vague to powerful. Master these and you'll consistently get results you can actually use — from any AI, for any task.

The framework

The 5 elements of a great prompt

Every strong prompt has some combination of these. Most weak prompts are missing most of them.

R

ROLE

Who should the AI be?

C

CONTEXT

What's the background?

T

TASK

What do you need done?

C

CONSTRAINTS

Any rules or limits?

F

FORMAT

How should the output look?

Now try it on your own prompt.

Paste any prompt below — we'll score it and explain exactly what's missing.

Your prompt

write me a cold email to a startup founder
Fix this prompt →

Live Analysis

type a prompt first

← Type a prompt to analyse it

The quick cheat sheet

ElementAsk yourselfExample phrase
ROLEWho should the AI be?"You are a senior [profession] with expertise in [topic]..."
CONTEXTWhat's my situation?"I'm a [role] working on [project]. The goal is [outcome]..."
TASKWhat exactly do I need?"Write / List / Compare / Summarize / Draft [specific deliverable]"
CONSTRAINTSWhat should it avoid?"Keep it under 150 words. Avoid jargon. Tone: casual and direct."
FORMATHow should it look?"Respond in a numbered list / markdown table / JSON / email format."

Keep going

The prompt engineering library

Short reads, about 5 minutes each. Start with the basics, then pick what you need.

A tangled scribble turning into a neat stack of blocks

The basics

What is Prompt Engineering?Learn what prompt engineering is and why it matters for getting better results from AI tools.How to Use Role in AI PromptsDiscover how assigning a role to an AI model shapes its tone, expertise, and output quality.How to Add Context to AI PromptsLearn how providing background context in your prompts leads to more accurate and relevant AI responses.Defining the Task in Your AI PromptFind out how clearly stating the task in your prompt is the single biggest factor in AI output quality.Using Constraints in AI PromptsLearn how adding constraints to your prompts keeps AI output focused, appropriately scoped, and on-brand.Controlling Output Format in AI PromptsDiscover how to specify output format in your prompts to get bullet points, tables, code, or structured text on demand.Zero-Shot Prompting: Examples, vs Few-Shot & Full Guide [2026]Zero-shot prompting gives AI a task with no examples — and it works when structured correctly. Learn the definition, see real examples, and know when to switch to few-shot.Few-Shot Prompting: Examples That WorkLearn how including a few examples in your AI prompt dramatically improves output quality and consistency.Chain-of-Thought Prompting GuideExplore chain-of-thought prompting and how it helps AI models reason through complex problems step by step.System Prompts Explained: What They Are, Examples & How to Write OneSystem prompts are the invisible instructions that control AI behavior. Learn what a system prompt is, see real examples for ChatGPT and Claude, and how to write one.Prompt Chaining for Complex TasksDiscover how prompt chaining breaks large AI tasks into a sequence of focused prompts for better, more reliable results.Temperature and Creativity in AI ModelsUnderstand what temperature settings do in AI models and how to use them to control creativity vs. precision.Prompt Injection: What It Is and How to Avoid ItLearn what prompt injection attacks are, how they work, and the key strategies for defending against them.Best Practices for Writing AI PromptsA concise checklist of proven best practices for writing AI prompts that consistently produce high-quality results.How to Write Better AI Prompts (Step-by-Step)A step-by-step guide to writing AI prompts that get better results from ChatGPT, Claude, Gemini, and any other model.

Advanced techniques

Advanced Role Prompting TechniquesGo beyond 'act as' with layered role prompts that unlock sharper, more specialised AI responses.Meta-Prompting: Asking AI to Write PromptsUse AI to design better prompts for itself — a technique that dramatically improves output quality.How to Build Reusable Prompt TemplatesBuild a personal prompt library with reusable templates that save time and produce consistent results.Iterative Prompting: Refine as You GoTreat prompting as a dialogue — iterate and refine each response to reach exactly the output you need.Self-Consistency Prompting ExplainedImprove AI accuracy by generating multiple reasoning paths and selecting the most consistent answer.Tree of Thoughts Prompting GuideTree of Thoughts lets AI explore multiple reasoning branches before committing to a final answer.ReAct Prompting: Examples, Templates & Full Guide [2026]ReAct prompting makes AI think out loud before acting — the technique behind modern AI agents. Includes examples, templates, and when to use it.Maieutic Prompting: Socratic AI QuestioningUse Socratic questioning in your prompts to make AI explain its own reasoning until contradictions surface.Generated Knowledge PromptingAsk AI to generate relevant background knowledge before answering — it significantly boosts accuracy.Directional Stimulus PromptingAdd a hint or keyword stimulus to your prompt to steer AI output in a specific direction.Context Stuffing: Maximizing the Context WindowLearn how to use large context windows effectively — what to include, what to cut, and how to structure input.Controlling AI Output FormattingTell AI exactly how to format its output — tables, JSON, markdown, bullet lists — and get consistent results.Negative Prompting: What to Tell AI to AvoidExplicit exclusions in prompts prevent common AI failures — filler, hallucinations, and off-topic responses.Getting Structured JSON Outputs from AIReliably extract machine-readable JSON from AI models for use in apps, pipelines, and automations.How to Debug a Bad AI PromptDiagnose and fix underperforming AI prompts with a repeatable debugging process.

Fix common AI problems

Why ChatGPT Ignores Your Instructions (and How to Fix It)ChatGPT skipping rules, going over word limits or drifting off topic? Here's why it happens and seven fixes that make it follow instructions.Why ChatGPT Gives Generic Answers (and How to Get Specific Ones)Bland, could-apply-to-anyone answers from ChatGPT come from missing context. Here's what to add so the answer is about your situation.Why Your AI Image Doesn't Match Your Prompt (ChatGPT, Gemini, Midjourney)Wrong style, wrong layout, extra objects or garbled text? Here's why AI images miss what you asked for, and how to write image prompts that land.How to Prompt Google AI Mode for Better AnswersGoogle AI Mode answers whole questions, not keywords. Here's how to phrase searches so it gives you a specific, useful answer, with examples.Why the Same Prompt Gives Different Answers in ChatGPT and ClaudeSame prompt, different results across ChatGPT, Claude and Gemini, or even twice in the same chat. Here's why, and how to get consistent answers.

Reference

What is a prompt?The plain-English starting point.AI glossary100 AI and prompt terms, explained in one line each.Prompt buildersWhat a prompt builder does and how to pick one.Prompt researchWhat real prompts look like, in numbers.Why PromptItInHow it differs from templates and dictation tools.

Let PromptItIn apply all 5 elements for you

Describe what you want in plain words. PromptItIn asks what's missing, then writes the full prompt for ChatGPT, Claude or Gemini.

Try PromptItIn freeAdd to Chrome, free