Fix

"The conversation is too long, please start a new one"

ChatGPT stopped mid-project with this exact message. It's a hard limit, not a bug — and your work isn't gone. Here's what it means and how to keep going in a fresh chat without re-typing everything.

What the error actually means

Every time you send a message, ChatGPT re-reads the whole conversation — your messages, its replies, and anything you pasted in. All of it has to fit inside a fixed budget called the context window. When a thread grows past that budget, ChatGPT can't fit a new message alongside the full history, so it stops and shows this message.

It usually lands at the worst moment: a long research thread, a writing session, planning that has spanned days. Nothing broke. The chat simply got too big to continue in place, and the only way forward is a new one.

The wrong ways to deal with it

Two instincts feel obvious, and both waste your time:

Retyping it all from memory. Slow, and you always leave out the details that mattered — the constraints you had to repeat, the approach you already ruled out.
Pasting the entire old chat into a new one. This recreates the exact problem you just hit. A wall of old text fills the fresh budget immediately, and you're back at the limit within a few messages.

And no, ChatGPT's memory feature won't rescue you here. Memory remembers facts about you across chats — your name, your preferences. It does not carry the decisions and half-finished work inside this specific thread. You have to move those over deliberately.

The right way: write a handoff

Instead of dumping the whole history, give a new chat a tight briefing — just the parts that matter. Do it in four moves:

State the goal in one line. What are you producing? Be concrete: "a Q3 launch plan", "a rewrite of chapter 4", "a script that parses these logs".
List the locked decisions. The things you already agreed and don't want reopened — the structure, the tone, the approach you ruled out.
Write down the constraints you kept repeating. If you reminded ChatGPT of the same rule again and again ("keep it formal", "cite sources", "no bullet points"), those are the first things a fresh chat won't know. Make them explicit.
Capture where you left off. Paste the latest draft, or the exact question you were mid-answer on.

Then open a fresh chat and paste it as the very first message. Here's a template you can copy:

Pick up exactly where we left off. Everything below is settled — don't re-ask it.

GOAL: [what we're producing, in one line]

LOCKED DECISIONS:
- [decision 1]
- [decision 2]

CONSTRAINTS (keep these every time):
- [constraint 1]
- [constraint 2]

WHERE WE LEFT OFF:
[paste the latest draft, or the exact question we were on]

Continue from here.

A fresh chat is faster, not weaker

A saturated conversation spends most of its budget re-reading a giant history — that's why long chats feel slow and forgetful. A fresh chat with a clean briefing is quicker and sharper, and it stops contradicting decisions you already made. Restarting well is an upgrade, not a loss.

The catch with doing it by hand

Writing that briefing yourself works, but it's slow — and you tend to leave out the constraints you were most sick of repeating, which are exactly the ones that matter. That's the job Uncook automates.

The faster way

Let Uncook write the handoff for you

Free & unlimited. No account.

Paste a share link to your ChatGPT conversation. Uncook reads the whole thing and writes a clean reboot prompt — your goal, locked decisions, repeated constraints and current state, already assembled. Skim it, paste it into a fresh chat, and keep going.

Uncook my chat →

Honest about your data: a share link is fetched once through our server to read the conversation, and its content is sent once to our AI provider (Anthropic) to generate your handoff — then discarded, never stored, never used for training. A share link makes the chat viewable by anyone with the URL; un-share it once you're done.