A prompt that worked once and lives in someone's chat history is not a team asset. Documentation is what turns it into one: what the prompt is for, what inputs it needs, what good output looks like, and, most usefully, when not to use it. This prompt generates that documentation in a consistent shape, which is what makes a collection of twenty prompts read as a library rather than as twenty personal notes.
The sections readers actually use are the ones most people skip when writing documentation by hand: sample input, sample output, and known limitations. Fill those in with something real even if you abbreviate it, because a documented prompt with an invented example teaches the wrong expectations to everyone who copies it. When you review what comes back, read the "when not to use" section hardest. If the model could not name a case where the prompt fails, it probably did not understand the prompt, and it certainly cannot warn the next person away from misusing it.
Documentation is how a prompt survives the person who wrote it. The limitations section is the part that keeps the next person out of trouble.
Fill in your own details below; the prompt updates as you type. When it's ready, copy it into Claude or whatever AI tool you use.
Create standardized documentation for prompts used in IA workflows, ensuring prompts are reusable, maintainable, and shareable across teams.