This AI prompt helps you reply to an angry message without making it worse. Paste the message & say who sent it. It names the fear or need behind the tone & the problem they want solved, then gives you one calm line to open your reply.

It works on a text, an email or a work chat message.

Decode a tense message before you reply

Everybody has gotten that message that makes their chest tighten. The reflex is to fire back fast, which is exactly how it escalates. Before you reply, this prompt reads the message for the real problem hiding under the tone, then gives you a calm first line.

A message I got feels confrontational, and I want to understand it before I reply.

Their message: [paste it]
My relationship to them: [who they are]

Tell me:
1. The real fear or need driving their tone.
2. The actual problem they want solved.
3. One calm opening line I can use to lower the tension right away.

Why it works: A hostile message has 2 layers, the tone & the actual ask. Most people react to the tone & miss the ask. This pulls them apart, then gives you a first sentence that drops the temperature so you can deal with the real problem instead of the heat around it.

One gap: It trusts the message at face value. If the person is being vague on purpose, the need the model gives you is just a guess. It only sounds certain.

Simple fix: Add the backstory the model can’t see, what happened before this message landed, so it reads the situation & not just the words.

Want to go deeper: Once it decodes the message, paste back: “Now write me 2 replies. One that answers the real need head-on, one that buys me a day without sounding cold.” You get the read & the response in the same sitting.

Original post by u/EQ4C. Link

Before you send the reply

Read the reply once as the person getting it. If it still needs work, the email rewrite prompt does exactly that before it changes a word.

This prompt first ran in issue 5, next to a hard decision prompt that argues with you before you choose.