--- title: "TREF Framework: Task, Requirement, Expectation, Format" url: "https://fvivas.com/en/tref-framework-prompts-llm/" date: "2025-04-20" author: "Fabio Vivas" language: "en" description: "Create effective prompts with the TREF framework. A simple, fast, customizable structure for clearer and more useful results." categories: ["LLM Guide"] --- When creating prompts for AI, what’s often missing is a good balance of clarity, direction, and a clearly expected outcome. That’s exactly what the **TREF** framework delivers — a concise and practical structure to guide AI with precision, even for simple or one-off tasks. TREF is ideal when you know what you want but need to ensure the AI understands the essential criteria of your request — what needs to be done, which conditions to respect, what level of quality is expected, and how the content should be delivered. ## What Is the TREF Framework? The TREF framework is a simple and direct structure that guides the AI to perform a specific task according to your needs. Its four elements create focused, easy-to-follow prompts that deliver results you can use right away. Here’s how each part works: ### Framework Components - **Task:** Defines what you want the AI to do, like “create a plan” or “explain a concept.” - **Requirement:** Specifies key conditions or necessary details, such as “based on X” or “using simple examples.” - **Expectation:** Describes what kind of result you want, such as “a practical solution” or “a clear explanation.” - **Format:** Tells how the response should be presented — a “numbered list,” “table,” or “short paragraph.” Together, these components form a direct prompt that minimizes ambiguity and emphasizes what matters. ### When to Use It? TREF is perfect when you need something quick, practical, and well-organized. Use it if: - You want a task completed with specific conditions. - You need results aligned with a clear objective. - You’re looking for structured output to apply or share. - You want a fast but well-targeted response. 💡 **Use TREF when you want a ready-to-use, no-frills answer.**\ *Typical example:* “Do [Task], following [Requirement], to ensure [Expectation], in [Format].” ## Practical Examples of TREF in Action To show how TREF works, here are three real-world applications. See how it organizes requests clearly and effectively. ### Example 1: Social Media Content Plan **Context:** You’re a freelancer creating posts for a fitness client. ``` Create a social media content plan (Task), based on at-home workout tips (Requirement), to make it simple and motivational (Expectation), in a 5-day table including date and description (Format). ``` **Why it works:** The Task sets the goal, the Requirement defines the theme, the Expectation sets the tone, and the Format organizes the delivery. ### Example 2: Explaining Blockchain **Context:** You need to understand blockchain for a meeting, with no technical background. ``` Explain what blockchain is (Task), using simple language and a real-life example (Requirement), so it’s easy to understand (Expectation), in a short paragraph (Format). ``` **Why it works:** The Task is clear, the Requirement simplifies the concept, the Expectation sets the clarity level, and the Format keeps it brief. ### Example 3: Project Task List **Context:** You’re starting a design project and need to organize your work. ``` Create a task list for starting a design project (Task), considering a one-week timeline (Requirement), to make it practical and realistic (Expectation), in a numbered list with 4 items (Format). ``` **Why it works:** The Task states the need, the Requirement adds the timeframe, the Expectation ensures feasibility, and the Format organizes the output. ## Tips to Get the Most Out of TREF TREF is practical and straightforward, but fine-tuning each part can make it even more effective. Here’s how to adapt it to your specific goals. ### Customize for Your Objective Adjust the TREF components to reflect exactly what you need: - **Task:** Be precise. “Write something” is vague; “Generate 3 title ideas” is clear. - **Requirement:** Include what matters. “About marketing” is broad; “based on social media trends” is focused. - **Expectation:** Set tone or complexity. “Useful” is vague; “practical for beginners” is specific. - **Format:** Be intentional. Lists are fast, tables compare, paragraphs explain — think about your final use. **Quick Example:**\ “Suggest workshop themes (Task), using 2025 tech trends (Requirement), to make it innovative and relevant (Expectation), in a list of 3 short items (Format).” These adjustments make TREF flexible and perfect for any situation — from the simplest to the most specific. ## Start Using TREF Today With the **TREF – Task, Requirement, Expectation, Format** framework, you have a practical structure for creating objective, well-organized prompts. Whether planning, explaining, or brainstorming, it helps the AI deliver clear, usable results — saving you time and effort. 🎯 **Quick Summary:**\ TREF tells the AI **what to do** (Task), **under which conditions** (Requirement), **what the output should be like** (Expectation), and **in what form** (Format). 🔗 **Want to explore more frameworks like this?**\ Check out the [**Practical Guide to Prompt Techniques, Frameworks, and Formulas for LLMs**](https://fvivas.com/en/prompt-techniques-frameworks-formulas-llms/), with dozens of detailed and applicable structures for different contexts and goals, plus techniques and prompt engineering tips. 📘 **Bonus tip:**\ Download the free eBook **[“Prompt Engineering Unveiled”](https://fvivas.com/en/ebook-prompt-engineering-unveiled/)**, featuring easy explanations, practical examples, and strategies ranging from basic to advanced to master communication with AI.