When a long chat forgets your rules, ask for a handoff brief
In a long chat, an AI can start ignoring a rule you gave in the first message. The rule is fine; the chat is too long. Anthropic's docs say that as the token count grows, accuracy and recall degrade, and they call it context rot. Ask the AI for a short brief of your rules and decisions, then paste it into a new chat.
Why it matters
A rule from message one ("keep the trip under 500 dollars") gets broken two hours later.
Reminding the AI works for a few messages, then it slips again.
A new chat that starts with a short brief of your rules follows them again.
Do this
- When the AI breaks a rule you gave early on, stop reminding it.
- Send the line below in the old chat.
- Read the brief. Add any rule it left out.
- Open a new chat and paste the brief as your first message.
A line to send in the long chat
Write a handoff brief for a new chat: every rule I gave you, and what we decided so far. Keep it short and use a list.
One thing to keep in mind: The brief is all the new chat knows. Read it before you paste: a rule it leaves out is gone.
The docs are written for Claude; the same handoff works in other AI chats. Source: Anthropic: Context windows, "How the context window works" ("As token count grows, accuracy and recall degrade, a phenomenon known as context rot").
Ask me anything about AI and cloudA paid call on your own setup: your agents, your bill, your plan.Book a call →