Keep memory across sessions
Stop re-explaining your project, your preferences, and prior decisions every time you open a new Claude conversation.
Starting each Claude session from zero is a tax on everyone — you spend the first ten minutes briefing, and even then Claude misses things you've told it before. The fix is a lightweight memory layer: either a structured CLAUDE.md that travels with the project, or an external memory engine that Claude reads automatically.
Recommended
The 2 tools we'd reach for first
Simplest possible memory: a single `memory.md` Claude reads at the start of every session. Good enough for most solo and small-team use cases.
Multi-file memory (decisions.md, preferences.md, glossary.md) when a single file gets unwieldy. Still zero external dependencies.
Alternatives
Use these when your stack calls for them
External memory engine with search and tagging. Worth the setup if you're running dozens of active projects and need to search across them.
Prescribes a folder layout for context files across all your projects. Pair this with either pattern above.
Not sure where to start?
Browse all 409 tools in the full catalog, or explore by role or trust tier.
Browse catalog · By role · By trust tier