What You Need to Know
A conversation holds a finite working space. Everything said, attached and produced so far is what Claude is working with. Very long conversations approach the platform's length and usage limits, and longer ones consume more of the usage allowance. Separately, claude.ai has memory across chats. Chat search is prompted retrieval, where you ask Claude to find something in past conversations.
A greengrocer has planned the shop's Christmas veg boxes in one very long chat since August. It holds supplier choices, three abandoned box designs, the agreed contents and a much-revised price list.
When a conversation reaches its length limit, the official guidance is to start a new conversation, or use a Project to work with larger amounts of information. Treat it like a long meeting: know it will end, and decide what must survive it.
The Three Moves
Restart when a piece of work is done, when the chat holds dead ends and superseded drafts, or when it nears its limits. A marathon chat carries everything it has ever seen, the greengrocer's three abandoned designs included. A clean start with exactly the right context beats that.
Summarise before you leave: have Claude summarise the decisions made, the current state and what remains. Then verify it yourself against the source records. The summary inherits every uncorrected error in the chat it summarises. The greengrocer checks hers against supplier emails and invoices. The verified summary becomes the new chat's opening brief.
Persist any specific working output the next conversation will need. Paste it in, re-attach the file, or store it in Project knowledge, where every chat in the Project starts with it. A standing document that keeps being revised and shared has a fourth home: an artifact, iterated as one evolving version. The greengrocer pastes in the agreed contents, makes the staff's weekly price list an artifact and stores the supplier list in Project knowledge.
What Memory Carries, and What It Does Not
Memory is built as a set of individual topics: your role, your projects, your communication preferences, your ongoing work context. It is context about you, not transcripts, drafts or attached documents. Each Project has its own memory. Inside a Project, Claude also uses your memory from chats outside Projects, but never another Project's memory. So memory saves you re-introducing yourself. It does not reliably transport an approved outline or a data file into the next chat. The greengrocer's new chat may know her shop, but its price list arrives only because she persisted it.
Two stale claims fail here, in opposite directions: that Claude cannot remember anything across chats, and that memory can be trusted to carry a working document. Explicit persistence remains the dependable pattern for a specific working output.
In the exam
This objective tests planning, not damage control, and the exam expects precision about what memory holds. Read the stem's tense. A planning stem describes work organised for what comes next: a long piece of work approaching limits, a project entering its final phase, a handover being planned. The greengrocer's case is one. A chat already producing poor outputs, diagnosed after the fact, is Domain 7's territory. The same marathon chat can appear in both kinds of item; the keyed answer follows the question asked. Here it is the proactive discipline: restart deliberately, summarise with verification (Domain 2 does not switch off at handover), persist what the next conversation depends on.
Nothing in the official guidance claims the product degrades in long chats; no keyed answer relies on it. Rollout dates and plan availability are dated facts and live on this domain's quick reference.
The continuity illusion
"A new conversation would lose all our context" is the marathon chat's best argument, and it is false. A verified summary and persisted documents carry forward the context worth keeping, at a fraction of the length and the usage.
Exam Traps
Relying on memory or chat search to carry a specific working output into a new chat
Memory holds context about you as individual topics, not transcripts, drafts or attachments, and chat search is prompted retrieval. A working output the next conversation depends on travels by explicit persistence: paste it, re-attach it, or store it in Project knowledge.
Pasting an unverified summary into the new conversation as the efficient handover
The summary inherits every uncorrected error and stale decision in the chat it summarises. Verification against the source records is the step that makes the handover safe, and the keyed answer always includes it; Domain 2 does not switch off at handover.
Keeping everything in one enormous chat because a restart would lose the context
The marathon posture dressed as continuity. The chat fills with superseded drafts and dead ends alongside the decisions that matter, approaches length limits, and consumes more usage; summarising and persisting carry forward everything worth keeping.
Practice Scenario
A wedding coordinator has run one enormous conversation across three months of planning: venue decisions, the running order, the readings, corrections to names. The final month starts now, the chat is thick with superseded detail, and the platform length limits are in sight. Some names and dates were corrected more than once, so any summary of the old chat may carry an outdated version. What is the best plan?
Build Exercise
Run the Handover Discipline on Your Longest Chat
What you'll learn
- How to extract a decisions-state-and-remainder summary from a long conversation
- How to verify a summary against source records before trusting it
- How to open a fresh conversation from a verified brief
- How to sort working outputs into paste-forward, artifact and Project knowledge
- How to test what memory does and does not carry into a new chat
- Find your longest-running Claude conversation and ask Claude for a summary of the decisions made, the current state, and the open questions
Why: The summary is the handover document; a long chat holds decisions worth keeping buried in detail that is not.
You should see: A structured summary in three parts: decisions, current state, open questions.
- Verify the summary against your source records, the emails, bookings or documents the decisions came from, and correct what it gets wrong
Why: The summary inherits every uncorrected error in the chat; verification is what makes it safe to build on.
You should see: A corrected summary, with at least one fix or a note that none was needed.
- Start a fresh conversation opened with the verified summary, and note what the clean start changes about the first few responses
Why: Experiencing the difference between a clean brief and a marathon chat makes the restart decision concrete.
You should see: Responses working from exactly the context you chose to carry, with nothing superseded in the way.
- Sort the old chat working outputs into their homes: what gets pasted forward once, what becomes an artifact because it keeps being revised and shared, and what belongs in Project knowledge because every future chat needs it
Why: Persistence is a routing decision; each destination serves a different life-span of output.
You should see: Each significant output assigned one home, with the artifacts created and the Project knowledge stored.
- Check what memory contributes at the fresh start: note what Claude usefully knows about you and your project, and confirm the working outputs themselves arrived only because you persisted them
Why: Seeing the boundary in your own work fixes the distinction the exam tests: memory carries context about you, persistence carries the work.
You should see: A two-line note: what memory supplied, and what explicit persistence supplied.
Sources
- How do usage and length limits work? — Anthropic
- Use Claude's chat search and memory to build on previous context — Anthropic
- Bringing memory to Claude — Anthropic
- What are projects? — Anthropic
- What are artifacts and how do I use them? — Anthropic
- Claude Certified Associate – Foundations Exam Guide — Anthropic