Memory
By default, each Claude conversation starts fresh; it doesn't remember your last chat. Memory features change that, letting Claude carry useful facts and preferences across conversations so you repeat yourself less.
What memory does
When enabled, Claude can retain things like:
- Who you are and what you do: your role, team, and goals.
- Your preferences: preferred tone, format, level of detail.
- Ongoing context: projects you're working on, recurring constraints.
The result: less re-explaining, more continuity. Claude starts to feel like a colleague who actually knows you.
Memory vs. Projects
These are complementary:
| Memory | Projects |
|---|---|
| Travels with you across all chats | Scoped to one body of work |
| Personal preferences & facts | Shared knowledge & instructions |
| Builds up gradually | Set up deliberately |
Use memory for "how I like to work," and Projects for "the context of this specific work."
Using it well
- Tell Claude what to remember. "Remember that I prefer concise answers and British English."
- Review and edit. Memory settings let you see and remove what Claude has stored. Keep it accurate and current.
- Correct stale info. If your role or preferences change, update or clear outdated memories.
Control and privacy
Memory is something you control:
- You can typically turn it on or off, and use modes (like a temporary/incognito chat) where nothing is remembered.
- You can view, edit, and delete stored memories.
- For sensitive one-offs, use a chat that doesn't write to memory.
Try it
Tell Claude one durable preference, e.g. "Always give me a TL;DR at the top of long answers." Then start a new chat and see whether it carries over.
Check availability
Memory features and their exact behavior vary by plan and roll out over time. Confirm current details and controls in your settings and the official docs.
You've got the coworking toolkit. Now let's put it to work on real tasks: Workflows for work.