Unabyss
MCP-native self-updating context layer for your AI
2.4K followers
MCP-native self-updating context layer for your AI
2.4K followers
Set it up once and never re-explain yourself to AI again. Connect the apps you use daily - Unabyss will extract, structure, and update your context automatically. Share it with any AI tool via MCP, with granular control over what each tool can see.
This is the 2nd launch from Unabyss. View more
Unabyss for Claude
Launched this week
Claude doesn't know what happens in GPT. Neither one really knows who you are or what your company does. Now they can.
Unabyss gives Claude memories from your other AI agents and everyday apps: email, Drive, GitHub, Notion, meeting recorders, and 20+ more. It saves new memories too, so GPT and Cursor stay in sync with the exact same context - sharper than wiring each tool into Claude one by one.
Finally, a real memory that follows you. Private. Portable.
Free Options
Launch Team / Built With
The portable memory angle is the strongest part of this. One thing I haven't seen addressed though ,what happens when you disconnect a source? If I connected Notion six months ago and Unabyss built context from it, then I disconnect Notion or delete those docs, does that memory get cleaned up automatically or does stale context from a deleted source keep surfacing in other tools indefinitely?
Unabyss
@jasnoor_singh_oberoi Hi Jasnoor - when you disconnect the app, there are two options: one to disconnect only, and the second to remove all the data associated with it.
@dominik_bartosik The two option approach also solves a trust problem, if disconnecting always deleted everything automatically, people would hesitate to connect sensitive sources in the first place knowing one accidental disconnect wipes months of structured context. Keeping those as separate actions is the right call.
How do you ensure that Claude & ChatGPT have the same memory? And does it work for OpenClaw too?
Unabyss
@amanda_baldwin1
Everything happens via our "store" tool in memory. You can save every conversation to Unabyss - just call it manually (e.g. "save it to Unabyss") or adjust your e.g. Claude preferences so that every chat is saved to your memory.
Some users choose the 1st option, some the latter - depending on how much information they want to transfer to Unabyss.
And yes, it works for OpenClaw too! In general, it works with every AI tool that has MCP :)
Unabyss
@amanda_baldwin1 when you are onboarding a new agent to Unabyss, there is a short setup/onboring flow that allows you to pick what you want to save. Based on that, you get global instructions that you can paste in your preferences - this way, context always flows between agents effortlessly.
Voquill
Nice idea. Congratulations!
How easy is it to clean up or remove memories if something gets saved by mistake?
Unabyss
@henry_habib extremely easy :)
if something gets saved by mistake, you can ask Unabyss to edit/remove it wherever you use it (Claude, GPT, Cursor, or in Unabyss itself :))
and you want to know what Unabyss knows about specific topic - you can talk it through. And (on Monday) we'll release Context View, where you'll see every information that Unabyss has!
Unabyss
@henry_habib super easy, there are 2 main ways in which you can do it: talk to our context agent in the app or do the same in your Claude, ChatGPT, or Hermes. Once you adjust it, it will stay fixed in your context.
The Eureka Database
Congrats on the second launch, Philip and team! I’ve definitely burned way too many tokens re-explaining the same product, preferences, and project context every time I move between Claude, Cursor, and ChatGPT, so this immediately feels useful.
The only thing I’d worry about is context getting bloated over time. Would love a simple way to see exactly what memory is being pulled in, how many tokens it uses, and quickly pin, trim, or exclude certain context before it gets passed to another tool.
Unabyss
@jeremy_galang thanks for the comment & feedback!
When it comes to the problems you described - we're actually working on showing exactly what's in your memory in a clear way. Token usage - that's interesting and we're currently researching that - just to check how much tokens are saved thanks to Unabyss. I believe that eventually you save them rather than burn tokens with additional context delivered via MCP.
Would love to see a quick preview of exactly what context each AI tool will receive before sharing, like a diff view so I can tweak it without digging through settings every time.
Unabyss
@dursunltxo preview the whole context; it's tricky because there is a lot of it. But there is a simpler solution that we implemented - a permission layer. When you connect to the agent, you can choose what type of information and from which app the agent will get it.
DiffSense
Hey guys! Quesstion. There are many shared memory apps out there. What sets this appart from the rest? Also curious about the name, its great! What was the reasoning behind the name? Thanks 🙏
Unabyss
@conduit_design first of all, we want to work in the background. Once you set up, there is no need to change how you interact with any agent. Unabyss does its thing, and you don't even know about it - you just see better results from agents and fewer questions asked.
About the name (thanks!), it comes from a combination of 2 words: "un" + "abyss" - as you can see in the screen below, when you hover over the logo in the footer, you will see the full etymology.
DiffSense
@dominik_bartosik Nice easter egg! I sent you some feedback on LI. Great product launch go for gold! 🚀 still nr1
I'm a designer, can I use it for multiple clients branding guidelines and other context?
Unabyss
@aleksandra_dabrowska02 yes! you can connect it to our Notion board with the client list, Gmail, and Calendar to get the full context, and then use it in Claude to get whatever you need!