Skip to content

When a prompt can't do the job

Each playbook names a gap rewording does not reach, and the smallest fix that is not a better prompt.

A prompt can score well and still be impossible to run, because it asks a model for something that model cannot do on its own: writing to a specific folder, running on a schedule, remembering last week. Rewording does not reach that.

Not sure which one you have?

Run the prompt through the prompt check first. If wording is the problem, it shows the weak spots. If the job needs something a prompt cannot give, it says so and links a playbook when one fits.