Create beginner Works with: Any AI

Negative Instructions to Positive Instructions Rewrite

Before after Has example

Replace don'ts with clear dos.

The Open buttons send the prompt text as currently shown below to that site only when you click. Fill in the placeholders first.

Rewrite my prompt's negative instructions as positive, testable instructions. My prompt: [PASTE_PROMPT]. The behavior I want: [DESIRED_BEHAVIOR]. Problems I saw in the output: [OBSERVED_PROBLEMS].

Return:
BEFORE: the list of negative instructions found in my prompt ("don't be verbose", "never use jargon", "don't make things up"), each marked with the likely problem ([vague], [conflicts with another rule], [no alternative given], [untestable]).
AFTER: for each, a positive version with a measurable standard. Examples: "Don't be verbose" -> "Answer in at most 120 words"; "Don't use jargon" -> "Use words a high school student knows; define any technical term in brackets"; "Don't make things up" -> "Use only the provided text; if something is missing, write 'not stated'."
KEPT NEGATIVES: a short list where a prohibition is appropriate (safety, privacy, legal), and how to phrase them clearly and narrowly.
REWRITTEN PROMPT: the full cleaned prompt, with rules grouped logically and ordered by importance.
TEST: three inputs to try, the expected differences, and a rubric.

Explain why positive instructions help (they describe the target, and they are easier to check), and warn about stacking too many rules. Preserve my original intent and any required safety constraints.

Fill in before use

Optional: type your details and the prompt above updates. Nothing is saved or sent.

Replace each bracketed placeholder with your own details, then paste into Any AI or another assistant.

BEFORE
"Don't be verbose." [vague] -> "Answer in at most 120 words."
"Don't use jargon." [untestable] -> "Use words a high school student knows; define any technical term in brackets."
"Don't make things up." -> "Use only the provided text; if something is missing, write 'not stated'."
Kept negatives: "Never include personal data in the output." (privacy, narrow and clear).
Rewritten prompt groups rules by importance. Test: three inputs, rubric for length and sourcing, then compare old and new outputs side by side. Too many rules reduce quality, so group them by importance and keep only the ones you can check. Preserve any required safety constraints exactly, and do not turn a narrow prohibition into a vague one.

• instruction-style