The CLEAR Prompt Framework

Introduction

CLEAR is a five-part checklist for writing better AI prompts: Concise, Logical, Explicit, Adaptive, and Reflective. It was proposed in 2023 by Leo S. Lo, a librarian and researcher, as a framework to help people work with AI tools thoughtfully. The first three letters shape the prompt you write; the last two shape what you do after the model responds. That makes CLEAR one of the few frameworks that builds revision into the method itself.

What you will build or practice

You will apply each CLEAR principle to a realistic prompt, then combine all five in one worked example you can adapt for your own tasks.

Before you begin

You need any AI chat assistant. If you have not yet learned the basic parts of a prompt, start with Prompt Engineering Basics — CLEAR builds on those elements and adds a process around them.

Key concept

The five principles in Lo's framework split into two phases. Concise, Logical, and Explicit are writing principles: cut clutter, order your instructions sensibly, and state exactly what you want. Adaptive and Reflective are working principles: adjust your prompt based on what comes back, and review what worked so your next prompt starts stronger.

Step 1: Concise — cut the clutter

Every unnecessary word dilutes the instruction. Drop apologies, filler, and backstory that does not change the task.

Wordy: I was wondering if maybe you could possibly help me come up
with some ideas for a birthday gift, if that is okay?
Concise: Suggest 5 birthday gifts under 30 dollars for a
60-year-old who loves gardening.

Step 2: Logical — order the instructions sensibly

Present things in the order the model should use them: role or context first, then the task, then the rules, then the material to work on. A prompt that scatters constraints before, inside, and after the source text is easy to misread.

Context: You are helping a small bakery write social posts.
Task: Write 3 short posts announcing weekend hours.
Rules: Friendly tone. Under 40 words each. No emojis.

Step 3: Explicit — say exactly what you want

Name the output format, length, audience, and scope instead of implying them. "Summarize this article" is implicit; explicit is better:

Summarize this article in exactly 3 bullet points for a busy manager.
Each bullet under 20 words. Only include findings, not methods.

Step 4: Adaptive and Reflective — refine after the response

These two principles kick in after the first answer. Adaptive means treating the response as feedback: if the output missed the mark, change your approach — rephrase, add an example, or split the task in two — rather than repeating the same prompt louder. Reflective means pausing after the task to evaluate the result critically: check facts against sources, notice what phrasing worked, and save it. Over weeks, reflection turns scattered attempts into a personal prompt library. The iteration loop in Prompt Engineering is a hands-on version of these two letters.

Practice exercise

Here is a full worked example applying all five letters. Study it, then build your own for a task from your week.

Context: You are an experienced hiring manager.
Task: Rewrite the job summary below for a junior data analyst posting.
Rules: Under 100 words. Plain language. List 3 must-have skills as
bullets. Audience is recent graduates.

[paste rough job summary here]

It is Concise (no filler), Logical (context, task, rules, material), and Explicit (length, format, audience). After running it, be Adaptive — if the tone is stiff, add one sample sentence in the voice you want — and Reflective: note which instruction mattered most and save the final version.

Common mistakes

• Treating CLEAR as only a writing checklist and skipping the Adaptive and Reflective steps.
• Confusing concise with short — a concise prompt can be long if every line earns its place.
• Repeating a failed prompt unchanged and hoping for a different answer.
• Never recording what worked, so every session starts from zero.

Check your understanding

1. What do the five letters of CLEAR stand for, and who proposed the framework?
2. Which letters apply before the model responds, and which after?
3. What is the difference between Adaptive and Reflective?

Key takeaways

• CLEAR stands for Concise, Logical, Explicit, Adaptive, Reflective — a framework from Leo S. Lo (2023).
• Write with the first three letters; work with the last two.
• Reflection is what turns one good prompt into a reusable skill.

Next steps

Practice the refinement loop in depth in Prompt Engineering, or see why these principles work in Prompt Engineering from First Principles.

Related resources

• Prompt Engineering Cheat Sheet
• Prompt Patterns Cheat Sheet
• Verify AI Answers

Newsletter or next lesson

Join the newsletter for a clear weekly lesson, or continue with the AI Beginner Roadmap.