Memory everywhere
Claude's memory now works everywhere, and you decide what's in it.
The memory in chat is the same memory Claude Cowork uses, so context carries between a conversation and a cloud task. It lives in Memory settings as short topic files you can read, edit, or delete. On by default for Free, Pro, and Max across web, desktop, and mobile; sensitive topics off until you opt in; Team and Enterprise admins gate it.
This is the independent field guide to all of it: chat and Cowork memory topics, Project memory, Claude Code CLAUDE.md, auto memory, and persistent context hygiene.
Signature desk
The Chips
Topic files on a blotter. Inspect, edit, or delete labeled synthetic memories. A sensitive chip stays off. A Team chip stays gated.
Synthetic specimen
The Chips
Seven labeled topic files on a blotter, none of them real private data. Inspect, rewrite, or throw one away. Dietary notes stay off until you opt in. Org writing voice waits on Team and Enterprise admin gating.
Original work by Claude Memories. The desk does not connect to a Claude account. Settings > Memory is still where the real list lives.
Sit at the deskDecision map
Choose the right memory surface
The fastest way to avoid memory confusion is to choose by scope before writing anything down.
Chat and Cowork
I want to know what changed in September 2026.
One memory across chat and Cowork, stored as editable topic files, with sensitive topics off by default.
Claude app
I want Claude to remember me across chats.
Turn memory on in Settings > Memory, read the topic list, and choose incognito for clean-room conversations.
Claude Projects
I want memory separated by client or initiative.
Use project memory boundaries, project instructions, and project knowledge deliberately.
Claude Code
I want Claude to remember how this repo works.
Use CLAUDE.md for human-written rules and auto memory for learned local notes.
Debugging
Claude is ignoring or misusing memory.
Check whether memory is enabled, loaded, specific, and free of conflicts.
Quotable
One memory, two surfaces.
Chat and Claude Cowork draw on the same editable topic files, so you brief Claude once.
Quotable
CLAUDE.md is context, not enforcement.
Use hooks or settings for hard requirements that must run regardless of model choice.
Quotable
Auto memory is useful only while it stays current.
Audit local Markdown memory after migrations, stale assumptions, and sensitive work.
Free tools
Build practical memory artifacts
Free client-side tool
CLAUDE.md Generator
Generate a concise, source-grounded CLAUDE.md template for Claude Code projects, with commands, conventions, memory guidance, and safety notes.
Use toolFree client-side tool
Memory File Designer
Design a persistent Claude memory scaffold for CLAUDE.md, CLAUDE.local.md, .claude/rules, and auto-memory notes without sending data to a server.
Use toolFree client-side tool
Context Budget Estimator
Estimate the token cost of CLAUDE.md and memory files, see what loads at startup, and spot oversized persistent context before it slows a session down.
Use toolGuide library
All pages
Memory everywhere
Claude Memory Everywhere: Chat, Cowork, and Who Controls It
Claude memory now spans chat and Claude Cowork as one editable set of topic files, with sensitive topics off by default and admin gating on Team and Enterprise.
Read guideStart here
What Is Claude Memory?
A plain-language map of Claude app memory, Claude Code memory, project memory, and developer-platform memory.
Read guideClaude app
Claude App Memory: Settings, Topics, and Controls
How Claude app memory works across web, desktop, and mobile, including memory topic files, Settings > Memory controls, incognito chats, and reset behavior.
Read guideProject boundaries
Claude Project Memory: Boundaries, Instructions, and RAG
How Claude project memory summaries differ from profile instructions, project instructions, and project knowledge retrieval.
Read guideTerminal memory
Claude Code Memory: The Working Map
The practical map of Claude Code memory: CLAUDE.md, CLAUDE.local.md, .claude/rules, auto memory, and /memory.
Read guideInstruction files
CLAUDE.md Guide: What to Put In It and Where
A practical, source-grounded guide to writing CLAUDE.md files for project instructions, local preferences, imports, and AGENTS.md bridges.
Read guideClaude-written notes
Claude Code Auto Memory: Storage, Limits, and Audits
How Claude Code auto memory works, where MEMORY.md lives, what loads at session start, and how to audit or clean it.
Read guideMaintenance
Claude Memory Hygiene: Keep Persistent Context Useful
A maintenance guide for keeping Claude memory concise, current, scoped, and safe across app memory and Claude Code memory.
Read guideSetup checklist
Claude Memory Setup Checklist
A practical setup checklist for Claude memory across app settings, Projects, CLAUDE.md, local memory, and auto-memory files.
Read guideDiagnostics
Claude Memory Troubleshooting
A diagnostic checklist for Claude memory not loading, Claude ignoring CLAUDE.md, stale auto memory, compaction surprises, and project boundary mistakes.
Read guideFreshness surface
Claude Memory Freshness Log
A dated update log for Claude memory behavior, source checks, and future changes to Claude app and Claude Code memory.
Read guide