A better brief. A useful experiment.

Turn a vague request into an instruction you can test.

Build a clearer brief, compare it with your original request and save a version. The five-point rubric checks stated fields, not model accuracy or writing quality.

Your improved prompt

Role / audience: 

Task: 

Permitted context: 

Constraints and checks: 

Output format:

Brief completeness: 0/5

  • ○ Role or audience is specified
  • ○ A concrete task is specified
  • ○ Source context is provided
  • ○ Limits and verification are stated
  • ○ The output format is specified

A field counts when it contains a minimum amount of text. This is a transparent structural check; it cannot judge truth, relevance, safety or output quality.

Find saved experiments →

Try the prompt

Compare the existing authored examples → · Choose an external tool →

The live pilot sends your prompt and the selected public lesson context to the configured AI provider. It requires sign-in and usage limits; conversations are not saved by this site.