Claude Memory is a new feature for team and enterprise users that remembers conversations to maintain context across projects. It creates separate, project-based memories and gives users full control to view, edit, or disable what Claude remembers.
And it's clearly built for the workplace (for now). The idea of having a separate memory for each project is a smart way to keep different workstreams from getting mixed up, like keeping product planning separate from client work. It also gives you a lot of control – you can see and edit exactly what Claude remembers in a summary, and there's an Incognito mode for chats you don't want saved.
It's rolling out to Team & Enterprise plans starting today. Hopefully, it will be available for Pro & Max subscribers soon.
Wow, finally a way for teams to actually keep context across projects! The control to edit or disable what Claude remembers is such a thoughtful touch—seriously solves so many “wait, what did we say last week?” moments.
Report
This is nice. that actually sounds super handy.
Report
Really interesting direction for Claude 🚀.
The project-based memory design feels like the right move context switching is one of the biggest productivity killers in teams, and having dedicated memories per project solves that in a very intuitive way.
I also like the transparency aspect (being able to view/edit summaries) that’s a big trust factor compared to black-box memory.
Curious though: how do you see this evolving for smaller teams or solo pros who juggle many contexts but may not be on enterprise plans yet? That feels like an audience that would benefit a ton.
Congrats on shipping this, excited to see where it goes 👏.
Report
It's really powerful. Enjoying Claude automatically generating these and being able to see and confirm its understanding of my work,
Report
When can we have it for non-teams users? Waiting for it!
this is my favourite feature in Claude and I'm regular user of it.
I would strongly suggest two things to improve it further-
1) Immediate chats (for e.g. referring to last 10 conversations) doesn't produce good results
2) Very old chats (more than 6 months old from my 410 chats) don't get referenced and the extended thinking gives bad results due to it
Report
Wish we could have this for individual accounts!
Report
Love It, started with using Cursor however now it feels like claude just gets me better, If you use git between all the bigger refactors, together with the MCP's it works amazing
Replies
Flowtica Scribe
Hi everyone!
Claude now has memories.
And it's clearly built for the workplace (for now). The idea of having a separate memory for each project is a smart way to keep different workstreams from getting mixed up, like keeping product planning separate from client work. It also gives you a lot of control – you can see and edit exactly what Claude remembers in a summary, and there's an Incognito mode for chats you don't want saved.
It's rolling out to Team & Enterprise plans starting today. Hopefully, it will be available for Pro & Max subscribers soon.
Fakeradar
@zaczuo What do you think, is Claude better or ChatGPT? We all witnessed an incredible technology race!
Agnes AI
Wow, finally a way for teams to actually keep context across projects! The control to edit or disable what Claude remembers is such a thoughtful touch—seriously solves so many “wait, what did we say last week?” moments.
This is nice. that actually sounds super handy.
Really interesting direction for Claude 🚀.
The project-based memory design feels like the right move context switching is one of the biggest productivity killers in teams, and having dedicated memories per project solves that in a very intuitive way.
I also like the transparency aspect (being able to view/edit summaries) that’s a big trust factor compared to black-box memory.
Curious though: how do you see this evolving for smaller teams or solo pros who juggle many contexts but may not be on enterprise plans yet? That feels like an audience that would benefit a ton.
Congrats on shipping this, excited to see where it goes 👏.
It's really powerful. Enjoying Claude automatically generating these and being able to see and confirm its understanding of my work,
Congratulations @kevin_garcia1 @kyleturman @scott_white_sf
@brooke_dewitt
this is my favourite feature in Claude and I'm regular user of it.
I would strongly suggest two things to improve it further-
1) Immediate chats (for e.g. referring to last 10 conversations) doesn't produce good results
2) Very old chats (more than 6 months old from my 410 chats) don't get referenced and the extended thinking gives bad results due to it
Wish we could have this for individual accounts!
Love It, started with using Cursor however now it feels like claude just gets me better, If you use git between all the bigger refactors, together with the MCP's it works amazing