Build
Peer-reviewedVerified 2026-09-01

CLEAR

Concise · Logical · Explicit · Adaptive · Reflective — Not a template — a rubric. Five principles for judging a prompt you have already written. This site’s quality meter scores against them.

More info

Use it for

Reviewing, teaching, and improving prompts. Use it after a template framework, not instead of one.

Avoid it for

Using it as a fill-in form. There are no slots — trying to write a prompt "in CLEAR" produces a prompt about prompting.

The slots

Each letter is a question the prompt has to answer.

Concise Optional process step

Is every sentence carrying weight?

Cut anything that does not change the output. Padding dilutes the instructions that matter, and OpenAI now explicitly recommends removing repeated instructions.

Logical Optional process step

Does it read in a sensible order?

Context before task, task before format, constraints last. Grouped instructions beat scattered ones.

Explicit Optional process step

Have you said it, or only implied it?

Name the format, the length, the audience, the exclusions. Every implication is a place where the model guesses.

Adaptive Optional process step

Can you change one thing and re-run?

Prompts are drafts. Vary one slot at a time so you learn which slot was doing the work.

Reflective Optional process step

Did you check the answer, or just accept it?

Verify claims, check the reasoning, and note what to change next time. Fluent output is not evidence of a correct answer.

Strengths

  • Peer-reviewed and citable, which matters in education and public-sector settings.
  • Framework-agnostic: it evaluates any prompt regardless of how it was built.
  • Reflective names the step almost everyone skips.

Weaknesses

  • Gives you nothing to fill in, so beginners find it hard to act on.
  • Written in 2023 for chat-era models; the "Concise" principle now competes with the detail that agentic tasks need.

Sources and grounding

  1. 1The CLEAR path: A framework for enhancing information literacy through prompt engineeringLo, L. S. (2023), The Journal of Academic Librarianship 49(4), 102720The peer-reviewed article that introduced CLEAR.
  2. 2The CLEAR Path (open-access record)University of New Mexico Digital RepositoryFreely readable version of the same paper.