Aider lives in your terminal and works a repo at a time. Its CONVENTIONS.md holds static rules, but everything you and Aider actually figure out together — the decision, the failed approach, the environment quirk — disappears when you exit. The next session, Aider re-reads the tree and re-derives context you already established, because there's nowhere durable for it to remember. This guide shows what persistent memory for Aider 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 Aider needs persistent memory
The problem isn't the model — it's statelessness. Aider 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 Aider stops re-learning your project and starts compounding what it knows.
What your Aider setup should remember
- Decisions and their reasons, distilled into short facts you can feed back into the next session.
- What approaches failed and why, so Aider doesn't re-attempt them.
- Repo-specific gotchas: flaky tests, build flags, environment setup.
- Conventions that go beyond what fits cleanly in CONVENTIONS.md.
What it changes, concretely
Before: each Aider session starts by re-explaining the parts of the project that aren't in the code. After: a small wrapper recalls the relevant memories and drops them into the chat as context — Aider opens already knowing the decisions you made last week.
How to add persistent memory to Aider
- Aider has no MCP, so call the Kit for AI REST API (or the typed SDK) from a small wrapper script.
- Before a session: recall(query) against the task, and pipe the top facts into Aider's context (e.g. a read-only file or a /add-ed note).
- After a session: remember(fact) the decisions and outcomes the session produced.
- Scope by repo with a namespace so each project's memory stays separate.
One memory across every tool
Because the memory lives in the cloud, it isn't locked to Aider. 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 Aider have persistent memory built in?
Aider 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 Aider long-term memory?
Call the Kit for AI REST API or typed SDK from your Aider setup: recall(query, end_user_id) to load relevant memories into context and remember(fact, end_user_id) to save durable facts — with an isolated namespace per user, so one integration serves all of them.
Related guides
Go deeper: the AI memory API · memory and RAG are different jobs — or read the pillar guide to persistent memory for AI.
Add persistent memory to your Aider agents in minutes — Start free and call remember and recall over the REST API or the typed SDK.