Prompt Comparison Tool
Score two prompt drafts on explicit task, context, output format, constraints and review instructions. See the gaps side by side.
Your result
Fill the fields and run the tool. Your input stays in this browser session.
Ready when you are.
What this tool does
Score two prompt drafts on explicit task, context, output format, constraints and review instructions. See the gaps side by side.
Example
Example: a one-line “write a report” prompt loses to a draft that identifies the source, reader, report sections and unsupported-claim checks.
How to use it
Paste both drafts for the same task. Compare checklist findings, improve the weaker draft and test both on the same input in your chosen model.
Useful for
Prompt iteration, team prompt reviews and testing a reusable instruction before rollout.
Limitations
Scores are simple text heuristics, not a measure of real output quality. Run an actual controlled test before making a decision.
FAQ
Does a higher score guarantee a better answer?
No. It shows whether common instruction elements are present, not whether the model follows them.
How do I compare results fairly?
Use the same model settings and source input, then assess outputs against a task-specific rubric.