Skip to content
Prompts for AI chat

Prompt enhancer

To use the prompt enhancer, open the Enhance tab, paste the prompt you already have and press Enhance prompt. Fixed rules then check which of seven fields your draft states. After that, they add a role and a quality bar, and they show the score before and after. Finally, you see the questions still open.

By The Promptcairn team (Product and testing). Last updated .

01 / Prompt enhancer
02 / Steps

How to use the prompt enhancer, step by step

  1. First, open the Enhance tab and paste your draft, up to 2,000 characters.
  2. Then press Enhance prompt and read the score before and after.
  3. Next, check the Already there and Added lists to see what the rules found.
  4. After that, read the Still open line, because it lists each missing field with its points.
  5. Finally, press Keep improving in the builder and answer those questions one at a time.

The first 3 runs each day are free. After that, see the pricing page for packs and plans, or read how we test the tool.

03 / Example

Prompt enhancer tested with a real input

We ran our own test input through this exact tool. The text and the numbers below come from that run, so you can repeat it and compare.

Prompt enhancer result preview from our test run
Output of our Prompt enhancer test run
Test input
"write a blog post about remote work"
Output
Enhanced prompt (rules only)
Score before
18
Score after
26
Added
Role, Quality bar
Still open
Audience, Output format, Context, Constraints, Example, Tone
Time to build
under 1 second
## ROLE
You are a content writer who knows search intent.

## TASK
Write a blog post about remote work.

## QUALITY BAR
- If a key detail is missing, state the assumption you made instead of guessing silently.
- Check the answer against every requirement before you finish.

How we test: every page has a fixed input, and we read each output by hand. More on the about page.

What the prompt enhancer checks in your draft

The rules look for seven fields. These are role, audience, output format, context, constraints, example and tone. Your own words stay in place as the task, so nothing you wrote is thrown away.

Next, the rules add two things. First, they add a role when the task implies one. For example, a landing page task gets a conversion copywriter. Then they add a fixed two-line quality bar. It asks the model to state its assumptions and to check the requirements.

Finally, you get a score out of 100 with fixed weights. The task counts for 18 points, and the audience counts for 16. Therefore, the same draft always gets the same score.

Why this AI prompt enhancer does not rewrite your text

This AI prompt enhancer is rule-based. So no model rewrites your prompt, and the result says that in plain words. We chose this because a rewrite can sound better while it guesses facts you never gave.

However, that honesty has a cost. The rules alone add few points, because they cannot know your reader or your goal. In contrast, the open questions carry most of the score. For instance, audience and output format are worth 30 points together.

So the largest gains come from your answers. The builder asks one short question at a time, and each answer becomes a line of the prompt. At 70 points, it tells you the prompt is good enough.

When a prompt enhancer is the wrong tool

First, skip it if you have no draft yet. In that case, start in the Build tab, which is free and unlimited. Enhance runs on the server, so it has a daily free limit and then needs credits.

It also checks text prompts only. For an image or a video prompt, the builder has its own fields, such as style, lighting and camera. Similarly, it cannot judge whether your facts are right. It only shows which parts of a prompt are present and which are still missing.

Fields and fixed weights behind the score

Fields and fixed weights behind the score
FieldPoints and how it is filled
Task18 points. Your pasted words are kept as the task
Specific detail8 points when the task has 8 or more words
Audience16 points. Detected, or left as an open question
Output format14 points. Detected, or left as an open question
Context12 points. Detected, or left as an open question
Constraints10 points. Detected, or left as an open question
Role8 points. Added by the rules when the task implies one of 14 task types
Example8 points. Detected, or left as an open question
Tone6 points. Detected, or left as an open question
Draft lengthUp to 2,000 characters
Questions

Prompt enhancer FAQ

Is the prompt enhancer free?

You get a daily free limit, and after that it needs credits. However, the Build and Ideas tabs stay free and unlimited.

Does the prompt enhancer use AI to rewrite my prompt?

No. Fixed rules detect fields, add a role and add a quality bar. So your wording stays as the task.

Why did my score only go up a little?

The rules can only add a role and the quality bar. Therefore, most points wait in the open questions, such as audience and output format. Answer them in the builder to raise the score.

Can I enhance a prompt for ChatGPT or Claude?

Yes. The result is a text prompt for a chat model. In the builder, you can also switch between Markdown headings for ChatGPT and XML tags for Claude.

What is a good prompt score?

The builder calls a prompt good enough at 70 out of 100. Still, the score only counts which fields are present, so read the prompt before you send it.

How long can my prompt be?

You can paste up to 2,000 characters. For a longer prompt, paste the core task first and then add the rest in the builder.

Sources and references