This AI prompt writes a reply to an email that gets you what you need. You paste the email & add one line: the goal of the reply. With a goal to aim at, the draft says something instead of sounding polite & empty.

Email reply prompt with a goal field

Paste an email into AI and say “write a reply.” The result sounds polite but doesn’t actually say anything. This prompt adds one field, the goal of the reply, that tells the model what you’re trying to accomplish.

Write a reply to this email.

Context: [paste the email you're replying to]

Goal of this reply: [what you want to accomplish, e.g., set a deadline, push back on scope, keep the relationship positive]

Rules: keep it direct, no filler, structure it as acknowledge then respond then next step.

Why it works: Without a goal, the model doesn’t know what you’re trying to do. It plays it safe and writes something polite that doesn’t really say anything. The goal field gives it a direction. You also don’t need a tone field here. When you paste the original email, the model picks up the tone from context.

One gap: The “acknowledge then respond then next step” structure is hardcoded. It works for pushing back on scope but breaks for other email types like delivering bad news or declining a request.

Simple fix: Change the structure line to “structure it in whatever order fits the goal” and let the model decide.

Original post by u/Rich_Specific_7165. Link

When the email is a hard one

If the email made you angry, the angry message prompt finds the real problem under the tone first. Already wrote your own draft? The email rewrite prompt checks it before you send. If the reply needs a real talk instead, use the hard conversation prompt.

This prompt first ran in issue 1.