The RTF Prompt Framework: Role, Task, Format Made Easy

If you've ever typed a question into an AI chatbot and gotten a vague, rambling, or off-target answer, the problem usually isn't the AI — it's the prompt. Most people write requests the way they'd fire off a quick text message. But AI tools respond best to clear structure, and one of the simplest structures you can learn is the RTF prompt framework.
RTF stands for Role, Task, Format. It's a three-part recipe you can apply to almost any request, whether you're drafting an email, summarizing a report, or planning a marketing campaign. Best of all, it requires zero coding and takes about 30 seconds to master.
What Is the RTF Prompt Framework?
The RTF framework breaks every good prompt into three ingredients:
- Role — who you want the AI to act as.
- Task — what you want it to do.
- Format — how you want the answer delivered.
When you give an AI all three, you eliminate the guesswork. Instead of the model choosing a random tone, depth, and layout, you tell it exactly what you need. The result is faster, more relevant answers with far less back-and-forth.
Why It Works
Modern AI models like ChatGPT, Claude, and Gemini are trained to follow instructions. But they can't read your mind. A vague prompt like "write about email marketing" could produce a tweet, a textbook chapter, or a sales pitch — the AI has no way to know which. RTF removes that ambiguity by defining the perspective, the goal, and the shape of the output up front.
Breaking Down Each Part
Role: Set the Perspective
The Role tells the AI what expertise or point of view to adopt. This single line dramatically changes the vocabulary, assumptions, and depth of the response. Compare "explain compound interest" with "Act as a financial advisor explaining compound interest to a first-time investor." The second version signals tone, audience, and expertise all at once.
Examples of useful roles:
- "Act as an experienced hiring manager."
- "You are a patient math tutor for a 10th-grade student."
- "Act as a small-business bookkeeper."
- "You are a professional copywriter specializing in local restaurants."
Task: Define the Job Clearly
The Task is the action you want performed. Be specific about the subject, the goal, and any constraints. Weak tasks are broad ("help with my resume"). Strong tasks are precise ("rewrite the work-experience section of my resume to emphasize leadership and measurable results").
Include details that matter: the topic, the audience, the length, the purpose, and anything the AI should avoid. The more relevant context you provide, the sharper the output.
Format: Shape the Output
The Format tells the AI how to structure its answer. This is the part most people forget, and it's often the difference between a wall of text and something you can use immediately.
Format instructions might include:
- "Give me a bulleted list of five items."
- "Write it as a table with two columns."
- "Keep it under 150 words."
- "Use a friendly, conversational tone with short paragraphs."
- "Format as a step-by-step numbered guide."
Putting It All Together: Real Examples
Here's how the three parts combine into complete, ready-to-use prompts.
Example 1: For a Small-Business Owner
Role: Act as a social media marketer for a local coffee shop.
Task: Write three Instagram captions promoting our new autumn pumpkin latte.
Format: Keep each caption under 40 words, include two relevant hashtags, and use a warm, inviting tone.
Combined, you'd type: "Act as a social media marketer for a local coffee shop. Write three Instagram captions promoting our new autumn pumpkin latte. Keep each caption under 40 words, include two relevant hashtags, and use a warm, inviting tone."
Example 2: For a Student
"Act as a patient biology tutor. Explain how photosynthesis works to a high school student who is struggling with the topic. Format the explanation as five short steps, and end with a one-sentence summary."
Example 3: For a Working Professional
"Act as a project manager. Draft an email updating my team about a two-week deadline extension on the client website launch. Keep it professional but reassuring, under 120 words, and end with a clear next step."
Notice how each prompt reads naturally — you don't need special syntax or symbols. You're simply writing a clear instruction that happens to cover all three bases.
Common Mistakes to Avoid
Even with a framework, a few habits can weaken your results:
- Skipping the format. Without it, you'll often get more text than you need. Always specify length and structure.
- Being too vague in the task. "Write about our product" gives the AI nothing to work with. Add the product name, audience, and purpose.
- Choosing a mismatched role. Asking a "casual friend" to write a legal disclaimer will produce the wrong tone. Match the role to the job.
- Overloading a single prompt. If you need several different outputs, break them into separate RTF prompts rather than cramming everything into one.
How to Practice the RTF Framework
The fastest way to internalize RTF is to use it on real tasks you already do. Before your next AI request, pause and jot down three quick notes: who should it act as, what exactly do you want, and how should the result look. Within a week, this becomes second nature.
Try running the same request twice — once with a plain prompt and once using RTF — and compare the results. The difference in clarity and usefulness is usually obvious, and it reinforces why structure matters.
When to Go Beyond RTF
RTF is an excellent starting point, but it isn't the only framework. For more complex work, you can add elements like examples of the output you want (sometimes called "few-shot" prompting) or extra context and constraints. Some people extend RTF into RTFC by adding "Context" as a fourth piece. Start with the basic three, though — for the vast majority of everyday requests, Role, Task, and Format are all you need to get consistently strong results.
As AI tools become part of daily work in 2026, the ability to write clear, structured prompts is quickly becoming a core professional skill. The RTF framework gives beginners a reliable foundation they can rely on across any AI platform, no technical background required.
Ready to build real AI skills? Join the September 2026 cohort at Class For Jobs. Explore AI for Everyday Use — a practical, no-code course to get more done with AI every day, live and instructor-led with career support, resume help, and job-placement assistance.









