A newsletter lives or dies on four moments: the subject line that decides whether it is opened, the hook that decides whether the first scroll continues, the body sections that deliver value, and the call to action that converts attention into a click. This pack gives you tested prompt structures for each, so instead of asking an AI to “write a newsletter” and getting bland filler, you brief it like an editor would — with niche, audience level, length, and tone all specified.
How it works
The library is grouped into categories — hooks, sections, CTAs, and subject lines. Filter to the stage you are writing or search by keyword, then copy a prompt and replace the bracketed placeholders such as [niche], [audience level], and [tone]. Each prompt encodes a clear role and a concrete output format so the model returns something usable rather than generic. Everything runs locally; the tool only copies prompt text to your clipboard, and nothing you type is uploaded.
Tips and examples
Run the A/B subject line prompt on every issue and pick from distinct angles — curiosity, benefit, and number-led — rather than three variations of the same line. For the hook, paste the single most interesting fact from your issue and let the prompt build the opening around it; readers stay for specifics, not throat-clearing. Keep CTAs to one ask per email, since a newsletter with three competing buttons converts worse than one with a single clear next step. Finally, give the model a sample of your own writing so the draft sounds like you and needs only light editing.