AI agent with a memory
An AI agent with a memory is one that keeps what it learns about you and your work, attaches it to your account rather than to one app, and applies it wherever you use the agent next.
An AI agent with a memory is one that keeps what it learns about you and your work, attaches that to your account rather than to a single app, and applies it the next time you use the agent anywhere.
The distinction that matters is not that it can recall the current conversation. Every chat window does that. It is that what you teach it in one place is still there in another, and next week: a convention, a correction, a fact about how your team works.
Three things follow from that, and they are what to check in any agent that claims a memory:
- What it keeps. In Dropstone, rules and facts, what it has studied, and short summaries of past work. See Rules and facts.
- Where it applies. Every signed-in surface of the same account: chat, the CLI, the SDK, and the editor. See Memory across chat, CLI, and SDK.
- How you see and remove it. Ask it what it remembers and it shows you the list; asking it to forget something removes it. See See what Dropstone has learned.
Dropstone is built this way rather than as an assistant with a memory bolted on later, which is why the same memory is the thing that ties its surfaces together.
Related articles
- Agent memoryAgent memory is what an agent stores between sessions and brings back when it is relevant: standing rules, facts, studied material, and short summaries of past conversations.
- Cross-surface memoryCross-surface memory is one memory attached to your account that every surface reads and writes, so a rule set in the terminal holds in chat and something learned in chat is available to code running through the SDK.
- How Dropstone remembers youDropstone keeps a persistent memory that follows your account across chat, the CLI, and the SDK. Correct it once and it should not need correcting again, anywhere you use it.
- How an agent learns as it worksDropstone is a self-learning agent. It records what it learns from corrections, from finished work, and from sources you ask it to study, and it uses that the next time without being told. Here is the loop.