One place for your notes, decisions, and workflows. For you, your AI tools, and your team | Local-first & open source OR Basic Memory Cloud
Austin, TX
Joined June 2025
- Tweets258
- Following286
- Followers166
- Likes196
After a year of work, we have released Basic Memory v1.0. Lots of changes, both in the local Open Source version and the hosted cloud service.
youtube.com/watch?v=p2Xpmk9b…
I've found every "cloud agent" setup to be more trouble than its worth so far, but I keep trying. I'm biased but Basic Memory does help because each agent can at least see the same context, so I can ask Claude/Codex/OtherAgent, etc about the same problem.
1
89
One of the tricks I learned from a mentor is to ask "What problem are we trying to solve here?..." when things go off the rails. Works with AI and IRL.
1
35
We wrote up how our agents show their work, including one catching an agent being fed sketchy instructions:
basicmemory.com/blog/agents-…
1
1
28
Then a contradiction-check agent flags facts that disagree, for you to review as a diff, with full version history and one-click undo.
Drift stops being a silent failure you hope a background pass catches. It’s dated, visible, and correctable. That’s the real win.
1
27
They worry notes drift, and a later session just trusts them.
Our answer starts with the record itself: facts in Basic Memory are time-aware. An observation can say when it held, so a stale “in progress” doesn’t rot into a lie. It ages out, and you can ask what was true on any date.
1
25
Memory also shouldn’t be married to one model. It’s just files you own, so our agents don’t have to run on Claude.
1
1
22
Putting the model in control of its own memory is cool. Letting a human step in is better, if they want to.
In Basic Memory that’s your choice. Let the agents run, or review every change before it lands. The model doesn’t decide that for you.
1
2
24
First, a nit: it’s not “dreaming.”
It’s a background job that reorganizes your notes and checks them against the transcript. That’s a good thing. But naming it after sleep hides what it is, a process you should be able to open up and inspect.
1
1
38
Recommended read, and basically a map of what we’ve been building. Basic Memory is agent memory as plain Markdown you own, and last week we shipped the agents that maintain it.
A few things we’d add 👇
Memory is what allows Claude to learn from experience. In a new Engineering Blog post, we share how “dreaming” helps Claude organize what it’s learned, correct its mistakes, and find what it needs more easily the next time. anthropic.com/research/compo…
If I do say so myself, I think our agents are pretty cool. They are really simple, just do things, and of course, they are just notes. You can even have your assistant manage them via Basic Memory Cloud MCP tools.
basicmemory.com/blog/agents-…
1
60
Now I want basic memory users to be able to sync their AI knowledge bases from local <-> cloud via Dropbox!
1
32
Tigris has been an excellent partner for Basic Memory. Always great to work with.
We started Tigris to build the storage layer we wished existed, and our customers trusted us with their hardest workloads. Today we announced Dropbox plans to acquire Tigris. Everything our customers have today stays. And we finally get to build this at the scale it deserves.
1
2
159
Other way around :). Basic Memory provides MCP tools for your agents to store info as plain markdown notes. It indexes, links, and creates semantic embeddings for them for fast search by the AI. And you can read and update them yourself. Works will all the agents/harnesses...
11
Try using Basic Memory as your shared knowledge base for you and your AI agents.
every startup uses some combination of slack, linear, granola, github, google calendar, a crm, etc
agents move across these more since they interact through MCP. the lines between each of these integrations is blurring. for example, you want to associate a bug in linear with a user who reported it in your crm. or you want meeting notes in granola to end up as a set of to-dos in slack.
this data shouldn't live in different products, but in a shared space where agents can move across without friction. ideally, the agent then builds triggers on its own to alert itself when important things change, and then surface stuff back to the team.
imo these shouldn't live as independent workflows and saas integrations
1
3
131
I biased, but I use basicmemory.com. They can share a memory knowledge base. People use this across claude, codex, etc. local or cloud. share with other people too.
1
10
When did people start using “alpha” as a noun in this context? Is this an x thing or does it also happen IRL?
1
46
For anyone who can't stand Notion, but wants a place to see, read, and share memory notes for people and AI, consider checking out basicmemory.com.
2
16