ChatGPT Email Reply Prompts: 19 Copy-Paste Examples

The one prompt to reply to any email in ChatGPT#

Paste the email, then tell ChatGPT your role, what the reply must do, the tone and the length. That is the whole recipe. Replying to a message, DM or text instead of an email? The same prompt works; swap the word “email” for “message”. Copy the template below, fill the brackets, and you get a draft to edit instead of a blank reply box.

Which prompt do you need? Pick your reply intent#

Almost every reply is one of eight jobs: accept, decline, ask for clarification, defer or buy time, acknowledge receipt, push back, schedule, or say thanks. Find yours in the table, paste the one-liner after the email, and jump to the matching section below for a full prompt, a sample reply and a tip. The holding-reply prompts (acknowledge and defer) share one section.

How do you feed the original email to ChatGPT?#

Every good reply prompt has two ingredients: the email you are replying to, and a clear instruction about how to reply. Skip the first and ChatGPT invents a context that is not yours. Skip the second and it guesses your intent, usually landing on bland agreement. Four habits cover it.

1. 1

   1. State your role and your intent first

   Open with one line: who you are and what the reply must do. “I am the project lead. I need to decline this scope change politely but firmly.” It tells ChatGPT the social shape of the reply before it reads a word of the email.

2. 2

   2. Paste the full email or thread under a clear marker

   Add “Here is the email I’m replying to:” and paste the message verbatim. In a long back-and-forth, paste everything you can see, oldest at the top, so the model does not repeat points already covered or contradict something agreed three replies ago.

3. 3

   3. Add your constraints: facts, tone, length, what to avoid

   Give it what it cannot infer: the date you can really meet, the number you will agree to, the one point you must not concede. Then set tone and length: “Under 80 words, friendly but not apologetic, no date before next Thursday.”

4. 4

   4. Read it as the recipient, then ask for one tweak

   Never send the first draft blind. Check that it answers everything they asked and claims nothing untrue, then refine in plain language: “make the opening less formal”, “cut it by half”. ChatGPT keeps the context, so each tweak builds on the last.

How do I reply to accept or agree to something?#

Acceptance replies go wrong two ways: so terse they read as cold, or so eager they give away leverage. Say yes clearly, restate the detail that matters, and set the next step. The restate instruction turns a vague “sounds good” into a reply the sender can act on without a follow-up.

If the yes comes with a condition, put the condition in the prompt. ChatGPT will weave it in without making it sound like a threat, which is the part people fumble when they write it themselves.

How do I decline an email politely without burning the bridge?#

A good decline does three things in order: appreciates the ask, says no clearly and briefly, and leaves the relationship intact, often with an alternative. Name those three beats in the prompt. The most important instruction is “don’t over-apologize”; without it, models pile on “so sorry, I wish I could” until the no sounds guilty and uncertain. For more variations, see these AI prompts to say no politely.

Declining an invitation has a slightly different shape. There is no alternative to offer, so the warmth carries it. Tell ChatGPT it is an invite and ask for light regret that does not tip into a sob story.

How do I reply asking for clarification?#

Some emails ask you to act before you have what you need. The right reply is a crisp request for the missing pieces, not a guess. Ask ChatGPT for short bullets and a collaborative tone. Tell it to ask only for what you genuinely need, because models pad clarification emails with extra questions to seem thorough, which slows the other person down.

When the email is confusing rather than just incomplete, ask ChatGPT to reflect back its understanding and check it. “So that I’m understanding correctly, you’d like X by Y” is more graceful than asking the sender to re-explain, and it often surfaces the misunderstanding without anyone admitting they were unclear.

How do I send a short holding reply (acknowledge or buy time)?#

Sometimes the honest reply is “I’ve seen this and I’ll get to it properly soon.” Silence makes the sender anxious; a holding reply buys time without the cost. Confirm you received it, name a realistic date for the full answer, and stop. Give ChatGPT a real date, or it writes “shortly” and leaves you on the hook for an undefined deadline. If nothing needs to happen next, ask for one or two lines and do not let it invent an action.

Can’t name a date yet? Ask ChatGPT to set the expectation that you will follow up with timing. It keeps the ball moving without committing you to a deadline you might miss.

For pure receipts, such as a document that arrived or an FYI, the reply is one line. The mistake is either ignoring these or writing a paragraph where a sentence does the job.

When nothing needs to happen next, such as a thank-you note or an FYI, ask for an even lighter touch so ChatGPT does not manufacture an action that is not there.

How do I push back or disagree without sounding defensive?#

Disagreement over email is high-stakes because tone is easy to misread, and your own first draft tends to leak whatever frustration you feel. A strong pushback states your position, gives one clear reason, and protects the relationship. Defensiveness shows up as over-explaining; firmness is one reason stated once. Ask for “calm and respectful, firm not defensive”, and supply the reason yourself.

When you are pushing back on feedback or a decision rather than a request, have ChatGPT acknowledge the other view before stating yours. Leading with “you’re wrong” closes the conversation; “I see why you’d land there, and here’s what I’m weighing” keeps it open.

How do I reply to schedule or propose times?#

A scheduling reply offers real options, sounds flexible and makes it easy to pick. Hand ChatGPT your actual availability. Ask for a generic scheduling reply and you get “let me know what works for you”, which bounces the work back; two or three concrete slots get a faster yes.

To reschedule, the reply has to acknowledge the change, take light responsibility and offer new times in one move. Tell ChatGPT it is a reschedule so it leads with the right beat.

How do I write a thank-you reply that doesn’t sound generic?#

“Thanks so much, really appreciate it!” says nothing. A thank-you that lands names the specific thing the person did and why it mattered. ChatGPT can do that if you give it the detail; the prompt’s job is to push it past generic gratitude.

Thanking someone for an opportunity you are turning down needs the gratitude kept sincere and separate from the no. Forced thanks reads as sarcasm; specific thanks reads as grace.

How do I control the tone of a reply?#

“Make it professional” is too blunt to steer anything. Name a direction the model can act on, ideally comparative: “warm and conversational, like writing to a colleague I like” or “firm and direct, no hedging, but not cold”. When you are unsure of the register, “match the tone of the email I’m replying to” is almost always safe: it avoids both common errors, being stiff with someone casual and breezy with someone formal.

Treat tone as a second pass. Get the content right first, then nudge: “make the opening less formal”, “soften the second paragraph, it reads as annoyed”. Because ChatGPT keeps the conversation context, small tweaks stack without re-pasting anything.

How do I add my own context and constraints?#

The email tells ChatGPT what was asked. It does not tell it what you know: the date you can deliver, the number you will agree to, the policy you must cite, the one thing you must not promise. A reply built without them is fluent and wrong. List constraints plainly as non-negotiable inputs: “I can deliver by the 14th, not sooner. Our minimum order is 50 units. Do not commit to onsite support.” Leave them out and the model fills the gaps with optimistic guesses.

Your voice is a kind of context too. To get replies that sound like you, paste two or three of your own emails and ask ChatGPT to match your greeting, sentence length and sign-off. That lasts for the chat. To keep it, set it once in Custom Instructions; this guide on custom instructions for ChatGPT email writing shows how.

What are the most common reply prompt mistakes?#

None of these are the model being bad. They are the prompt leaving out something the model needed.

• Pasting only the last message of a long thread. If the sender’s real question was three replies up, a one-message paste produces a confident answer to the wrong thing.
• Not stating your role. Without “I’m the vendor” or “I’m the customer”, the model guesses, and a reply written from the wrong side of the table is worse than none.
• Leaving out your real constraints. Without the true date, number or limit, ChatGPT invents a plausible one. Fluent and wrong is the hardest output to catch.
• Asking for “professional” with no direction. Name the feeling instead: warm, firm, neutral, apologetic.
• Sending the first draft unread. Models occasionally agree to things you didn’t authorize or miss a buried question. A 15-second read as the recipient catches almost all of it.
• Letting it over-apologize. Declines especially drift into guilt unless you add “don’t over-apologize”.
• Forgetting it has no memory of the sender. Unless it can read the relevant thread, ChatGPT doesn’t know what you promised this person last week. Anything that matters has to be in the prompt.

Can ChatGPT see your inbox, or send the reply for you?#

It depends on how you use it. A plain chat window sees only what you paste. ChatGPT also offers Gmail and Outlook connectors on paid plans, and OpenAI’s June 2026 release notes describe sending an email from a connected Gmail or Outlook account directly from a chat, in the web app. Third-party guides describe a one-message-at-a-time flow where you approve each send. Availability varies by plan, region and, for work accounts, administrator settings, so check OpenAI’s Help Center for what your account has before you rely on it.

Even with a connector, the pattern is the same: you start a chat, ask for a reply to a specific thread, and approve the result. That is a good fit for the occasional important email. It is a slow loop for forty replies a day.

The prompts in this guide work either way. With a connector, ChatGPT can fetch the thread itself, so you can skip the paste and say “Draft a reply to the latest email from [name]”. Without one, paste the thread as shown above.

What does the copy-paste workflow cost at volume?#

For one important email, the prompts above are a fair trade. The cost shows up across a full inbox. Replying one email at a time through a chat window means:

• Copy the thread, switch to ChatGPT, type your role, intent and constraints, paste, read, tweak.
• Copy the result, switch back to your inbox, paste it into the reply box, fix the formatting, re-check the recipient.
• Repeat for every message, because each reply starts from the context you supply.
• Re-establish your voice and facts in each new chat unless you have set them up in Custom Instructions.
• Accept the privacy trade-off of mail leaving your mailbox for a general-purpose tool.

Where does AI Emaily fit?#

AI Emaily is an AI-native email client that drafts replies in the thread, on your real mailbox, so there is nothing to paste. It works across Gmail, Outlook and Microsoft 365, iCloud and IMAP providers. You set your voice and details once in your Personal Context, plus per-client profiles, and drafts follow them.

It runs in three modes. In Manual, AI works when you ask. In Copilot, AI drafts the replies and you approve every send. Autopilot, available on the Autopilot and Team plans, handles routine mail within rules you set, with undo and an audit trail. The eight reply intents in this guide become one action on the open email instead of a round trip to a chat tab.

It is not free forever. Pro has a 7-day free trial and then costs $12.42 a month billed annually ($14.99 monthly). Use the prompts here to learn what a good reply looks like; if the copy-paste loop is the part that costs you time, start a 7-day free trial of AI Emaily and compare.

Conclusion: paste the thread, name the intent, read before you send#

ChatGPT is good at drafting replies, and the recipe is short: paste the original message, state your role and intent in one line, add your real constraints and the tone you want, then read the draft as the recipient and ask for one tweak. Match the prompt to the job (accept, decline, clarify, defer, acknowledge, push back, schedule or thank), keep your real reasons in the prompt, and never send the first draft blind.

For a deeper look at what ChatGPT can and cannot do with mail, see what ChatGPT can and can’t do with your email.