Prompt optimizer

Paste any prompt. The checker scores it against eight things good prompts have, instantly. Then the AI enhancer rewrites it and you see both versions scored side by side.

Your prompt

No prompt handy? Try one:

Optimize for

0/100

Before

Underspecified

0 words

Clear task verb missingStart with an explicit action: “Write…”, “Summarize…”, “Compare…”.
Background or context missingSay who it is for and why you need it; models fill gaps with generic guesses.
Role or perspective missingGive a perspective: “You are a senior UX writer…”. It sets vocabulary and depth.
Output format stated missingName the shape of the answer: table, bullets, JSON, 3 paragraphs.
Length or scope limit missingCap the length: “under 150 words”, “5 bullets”.
Constraints or rules missingAdd rules: what to avoid, what must be included, reading level.
Example or reference missingShow one example of the output you want. Few-shot beats description.
Specific, not vague missingReplace vague words (good, nice, interesting) with concrete criteria.

Enhance with AI

Sends your prompt to the AI helper and returns a rewrite plus a list of what changed.

What the eight checks look for

Check	Why it matters	Quick fix
Clear task	Without an explicit verb, the model guesses what “about remote work” should become.	“Write a 600-word blog post…”
Context	The model cannot see your business, your reader or your goal.	“For HR managers at 50-person companies deciding on a hybrid policy.”
Output format	The shape you want is rarely the default shape.	“A table with columns Idea, Cost, Effort.”
Constraints	Tells the model what to leave out, which is often the difference between generic and useful.	“Avoid clichés; no more than one statistic.”
Role	Sets expertise, vocabulary and assumptions.	“You are an employment lawyer…”
Length	“Short” means different things to different models.	“Under 150 words.”
Example	Showing beats describing: one example fixes tone and format at once.	“Here is a post I liked: …”
Specific wording	Vague adjectives like “good” or “interesting” give the model nothing to aim for.	Replace with criteria: “surprising to a beginner”.

The checker is a heuristic: it looks for signals in your wording, so it can be fooled and it cannot judge whether your facts are right. Use it as a checklist. For the reasoning behind each item, see the prompt engineering guide, and to see how long your prompt is in tokens, use the token counter.

Questions

How is the prompt score calculated?

Eight checks, weighted: a clear task verb (18), context (16), output format (16), constraints (12), role (10), length limit (10), an example (10), and specific wording (8). They mirror the advice in OpenAI’s and Anthropic’s prompting guides. The score runs in your browser, instantly, with no AI involved.

What does the AI optimizer change?

It keeps your intent and facts, makes the task explicit, adds [placeholders] for missing context, specifies format and length, adds constraints and removes vague words. It then lists what it changed so you learn the pattern.

Is a higher score always better?

Usually, but not always. A quick factual question does not need a role or an example. Use the checklist as a prompt to think, not a quota. The side-by-side view shows which checks the rewrite added.

Is this a prompt enhancer or a prompt checker?

Both. The scorecard is the checker: it tells you what is missing. The AI rewrite is the enhancer: it fills the gaps. You can use either on its own.

Is my prompt stored?

Your draft is saved in your browser so it survives a reload. It is only sent to the AI helper when you press the optimize button.