Zed's assistant is quick and unobtrusive, but its knowledge of your project resets with the workspace. There's no durable place for the decisions you make together, the conventions you want enforced, or the gotchas you hit — so you re-explain them, session after session, and the assistant never quite grows into your codebase. This guide shows what persistent memory for Zed looks like in practice — what to store, what it changes, and exactly how to wire it up. New to the idea? Start with the complete guide to persistent memory for AI.
Why Zed needs persistent memory
The problem isn't the model — it's statelessness. Zed is only as consistent as the context it's handed each session, and without a durable memory that context resets to zero every time. Persistent memory is a store outside the model that keeps the facts, decisions, and preferences you build up, and surfaces them by meaning when they're relevant again — so Zed stops re-learning your project and starts compounding what it knows.
What your Zed setup should remember
- Decisions and their reasons, so the assistant argues from your history, not from defaults.
- House conventions it should apply without being reminded.
- Environment and build quirks specific to this project.
- Cross-file and cross-repo facts that no single buffer can hold.
What it changes, concretely
Before: you re-brief Zed's assistant on the project every time you reopen it. After: it recalls the relevant decisions and conventions the moment they matter — the assistant feels like it's been on the project the whole time.
How to add persistent memory to Zed
- Register the Kit for AI MCP server as a context server in Zed's settings with your API key.
- The assistant gains remember and recall and uses them to persist and retrieve project knowledge.
- Scope memories per project so context stays relevant and isolated.
- Because it's cloud memory, the same knowledge is available in Zed, Cursor, or Claude Code — one brain, every editor.
One memory across every tool
Because the memory lives in the cloud, it isn't locked to Zed. A decision you save here is recalled by every other MCP-capable tool and by your own scripts over the API — switch models or editors and your memory follows you. That's the difference between a per-app memory silo and a real memory layer for AI, and it's why the same account also handles document ingestion and cited search.
FAQ
Does Zed have persistent memory built in?
Zed keeps context within a session and through static config files, but it has no durable, cross-session memory of the decisions and preferences you build up — that resets every time. Persistent memory adds a store outside the tool so those facts survive restarts and follow you across machines.
How do I give Zed long-term memory?
Add the Kit for AI MCP server to Zed with an API key — remember and recall become tools it calls on its own. It takes a couple of minutes, and the memory is shared with every other MCP-capable tool you use.
Related guides
Go deeper: the memory layer pattern · share one memory across AI models — or read the pillar guide to persistent memory for AI.
Add persistent memory to Zed in minutes — Start free, drop in the MCP server with your API key, and remember/recall become tools Zed calls on its own.