Prompt
In one line: What you type to an AI. The art of writing effective prompts is called prompt engineering.
What is Prompt?
A prompt is the input you send to an AI - your question, instruction, context, or combination of all three. Despite all the complexity underneath, this is your primary interface with LLMs. The quality of the prompt is often the single biggest factor in answer quality - more so than the choice of model.
Types of prompts
- User prompt - What you type in the chat box. The most familiar form.
- System prompt - An instruction set by the app or developer before your message, shaping the model's persona and constraints for the whole conversation. You rarely see these, but every polished AI product uses them.
- Few-shot examples - Sample input/output pairs included to show the model the exact format you want. Especially effective for classification, extraction, and structured data tasks.
- Documents / context - Text, PDFs, or pasted data you attach for the model to analyse. This is the foundation of RAG-style workflows.
- Chain-of-thought prefix - A phrase like 'think step by step' that triggers chain-of-thought reasoning before the final answer.
Weak vs strong prompt examples
| Type | Weak prompt | Strong prompt | Why it matters |
|---|---|---|---|
| Writing | Write a tweet | Write a 220-char tweet about our new pricing. Casual tone, no hashtags, one emoji at the start. Key message: Pro is $9.99/mo | Format, tone, constraints, goal - all explicit |
| Summarise | Summarise this | Summarise in 3 bullet points for a non-technical manager. Focus on business impact, not technical detail | Audience + format + focus |
| Code | Fix this bug | This Python function throws a KeyError on empty dicts. Identify the root cause and return a corrected version with an inline comment explaining the fix | Expected behaviour + output format |
| Analysis | What do you think? | List the three strongest arguments for and against this proposal. Use a table. Be impartial | Structure + framing prevent one-sided answers |
What makes a good prompt
The strongest prompts share five properties:
- Role - Tell the model who it's being: 'You are a senior data analyst...'
- Context - Give it the background it needs to answer well, not just the question alone.
- Format - Specify the output format explicitly: bullet list, table, JSON, numbered steps.
- Constraints - State what to avoid: 'no jargon', 'under 150 words', 'cite sources'.
- Goal - Make the end purpose clear: the model should know what 'done' looks like.
To go deeper, explore the prompt library with 31 ready-to-use templates, or read the guide to writing better AI prompts. The full craft of prompt design is covered in prompt engineering. Two FAQ answers cover the practical side: how to ask a good question and saving prompts for reuse.
Prompt example
If you are using AskAI.free, a practical way to understand prompt is to ask a model to explain it, then ask for a concrete example in your own workflow. For example: "Explain prompt for someone using AI to write, code, research, or create images."
This turns the term from a dictionary definition into a decision-making tool: you can see when it affects prompt quality, model choice, output reliability, privacy, cost, or how much context the AI can use.
Why Prompt matters
Prompt matters because it changes how you choose, prompt, compare or trust AI systems. If you understand this term, you can ask better questions, spot weak answers faster and choose the right model or tool for the job.
A common mistake is treating prompt as isolated jargon. It usually connects to nearby ideas like Prompt engineering and RAG (Retrieval-Augmented Generation), so check those next if you want the full picture.
Common mistake with Prompt
The most common mistake is using the term as a label without changing behavior. When prompt comes up, ask what action should change: the prompt, the model, the input length, the evidence you request, or the way you verify the answer.
See it in action - Ask any AI about prompt on AskAI.free.
Try it free →