A Unified Framework for an Effective Prompt

A fundamental advancement in prompt engineering is the realization that a prompt is not a monolithic question but a structured document composed of distinct components. Let's talk about a unified framework that allows you to create effective prompts.

Role

This component defines the "character" or the "persona" of the model should embody. It's the primary method for controlling the tone, style, and domain of expertise.

Assigning a role constrains the model's vast knowledge base. Instead of drawing from the entire internet, it prioritizes information and linguistic patterns associated with that role.

Impact on Output

• Tone & Style: Determines if the response is formal, casual, witty, technical, academic, or empathetic.
• Vocabulary: Influences the choice of words (e.g., a "legal expert" will use precise legal terminology, while a "5th-grade teacher" will use simplified language).
• Complexity: Sets the level of detail and sophistication of the explanation.

Examples of Use

• Technical: "You are a senior database administrator specializing in PostgreSQL."
• Creative: "You are a 1920s noir detective narrating a case."
• Educational: "You are a physics professor explaining quantum mechanics to a first-year university student."

Context

This background information provides the necessary situational awareness and factual grounding for the task. Without context, the model operates in a vacuum and is more prone to making up information (hallucinating).The model will prioritize using the information given in the prompt over its more generalized, and potentially outdated, training data.

Types of Context

• Conversation History: Including previous user queries and model responses to maintain a coherent dialogue.
• Documents: Pasting in text from articles, research papers, product manuals, or legal documents for the model to analyze or summarize.
• User Data: Providing user profiles, preferences, or past behavior to generate personalized recommendations.
• Real-time Information: Supplying data from an external API call (e.g., current weather, stock prices) to ensure the response is timely and accurate.
  
  

Task

Sometimes also called the Instruction or Directive, this is the core imperative of the prompt. It must be a clear, specific, and actionable command detailing what the model should do.

• Clarity is Key: Ambiguous instructions lead to generic or incorrect outputs. "Tell me about marketing" is a poor task, whereas "Generate a 5-step marketing plan for a new vegan cafe targeting millennials" is a strong one.
• Use of Direct Action Verbs: Starting the task with a precise verb removes ambiguity.
  • Generative Verbs: Create, Write, Generate, Compose.
  • Analytical Verbs: Analyze, Compare, Contrast, Critique, Summarize.
  • Transformative Verbs: Translate, Rewrite, Reformat, Convert.
• Task Decomposition: For complex requests, you can break down the task into a sequence of numbered steps within the same prompt (e.g., "1. Summarize the provided article. 2. Identify the three key arguments. 3. Write a counter-argument for each point.").
  

Examples

These demonstrate the desired input-output pattern, enabling a powerful form of in-context learning. This is often the most effective way to control the fine-grained details of the output. They are sometimes also called Shots. Hence, the terms One-Shot Prompting, or Few-Shot Prompting.

What Examples Teach

• Structure & Format: Showing the model an example in JSON teaches it to respond in JSON.
• Reasoning Process: For complex tasks like chain-of-thought, an example can show the model how to break down a problem step-by-step.
• Nuance & Style: Examples can demonstrate a specific tone or stylistic choice more effectively than a verbal description.
• Best Practices: Examples should be consistent, high-quality, and representative of the desired output. They should mirror the structure of the actual task, and cover as many cases as possible.
  
  

Constraints

These are the "guardrails" that define the rules or warnings for the boundaries of the response. They specify what the model should not do and are critical for safety, consistency, and brand alignment.

Common Constraints

• Length: "Do not exceed 300 words," "Provide a response in exactly three sentences."
• Content: "Do not provide medical or financial advice," "Avoid mentioning competitors," "Do not use profanity."
• Style: "Write in the active voice," "Do not use technical jargon," "Explain this as you would to a complete beginner."
• Information Sourcing: "Only use information from the provided context," "Do not use any knowledge you have about events after 2023."
• Output Format (or Structure): This explicitly defines the desired structure of the model's response, which is crucial for reliability and programmatic use. When an LLM's output is fed into another software system, a consistent and predictable structure is non-negotiable. Specifying the format turns the LLM into a reliable API component.
  

Delimiters

To help the model distinguish between these different components, it is a best practice to use clear delimiters. Structuring the prompt with markers like Markdown headers (e.g., ###Instruction), XML-like tags (e.g., <CONTEXT>...</CONTEXT>), or simple labels ensures the model understands the role of each piece of information you provide, dramatically increasing the reliability and quality of the final output.

How to use the framework

To help you create prompts that abide by this powerful framework, we have created a prompt that generates prompts.

**Your Goal:** You are a world-class Prompt Engineer. Your function is to take a user's simple, brief idea and transform it into a comprehensive, structured, and highly effective prompt that can be used to instruct another AI.

**Your Process:**
1.  Read the user's simple request (which will be provided at the end).
2.  Analyze the user's core intent. What are they trying to achieve?
3.  Based on their intent, construct a detailed prompt with the following components, using clear markdown headers (e.g., `###ROLE###`) as delimiters.
4.  Intelligently infer and add specific details for each component to make the prompt as effective as possible.

**Components to Include in Your Generated Prompt:**
* **`###ROLE###`**: Define the best possible persona or expert character for the AI to adopt to fulfill the user's goal. (e.g., "You are an expert travel agent," "You are a professional copywriter").
* **`###CONTEXT###`**: Provide necessary background information that the AI would need. You may need to invent plausible context based on the user's request. (e.g., "The user is planning a 7-day trip to Japan in the spring," "The product is a new brand of luxury soap").
* **`###TASK###`**: Write a clear, specific, and actionable instruction. Use strong action verbs and break down complex requests into numbered steps. This is the core of the prompt.
* **`###CONSTRAINTS###`**: Add a list of "rules" or boundaries for the AI to follow. These are things the AI *should not* do. (e.g., "Keep the response under 300 words," "Do not use technical jargon," "Avoid mentioning specific brand names").
* **`###EXAMPLES###`**: (Optional but Recommended) If the task is complex or requires a very specific style, provide a high-quality example of the desired input-output pattern.
* **`###OUTPUT FORMAT###`**: Explicitly define the structure of the final response. (e.g., "Provide the output as a JSON object," "Format the response as a markdown table," "Use a numbered list").


**Now, apply these instructions to the user request I provide below.**
{YOUR REQUIREMENTS HERE}

Describe your requirements in 2-3 sentences to receive a production-ready prompt for your next prompt engineering project. We encourage you to experiment with the output and share your findings in the comments below.