The prompt library,
explained.
How the prompts work, how to customize them, and how to use them safely.
Which AI tools do these prompts work with?
All of them. The prompts are tool-agnostic plain text, so they work with any capable chat assistant or API. Stronger, more recent models will follow the instructions more reliably, but nothing here is tied to a specific product.
What are the {{variables}} in the prompts?
Placeholders you replace with your own details before sending. For example, {{company}} or {{audience}}. Each prompt page lists every variable and what to put there. Swap them in, delete any you do not need, and run it.
Are these prompts free to use?
Yes. Copy, adapt, and use them freely in your own work. There is no sign-up and nothing is tracked beyond basic, privacy-respecting analytics.
How should I adapt a prompt?
Start from the closest prompt, fill in the variables, then refine: add the context the model is missing, tighten the output format, and iterate on the response. See the guide on how to write a good prompt for the habits that help most.
Should I paste sensitive data into a prompt?
Be careful. Avoid pasting secrets, personal data, or confidential information into third-party tools unless you know how that data is handled. If you need to share a document, consider stripping sensitive details first (the SDEN prompt anonymizer does this in your browser).