Maximem Vity for OpenClaw
One memory across OpenClaw, ChatGPT, Claude & Gemini
140 followers
One memory across OpenClaw, ChatGPT, Claude & Gemini
140 followers
MemoryPlugin gives you one persistent “brain” across OpenClaw, ChatGPT, Claude, and Gemini. Connect a single Chrome extension to inject the right context, search your past conversations, and sync Chrome and X/Twitter bookmarks. It also turns what you learn into WaitPro flashcards, so your best prompts, decisions, and research stay reusable across sessions and tools.
Free
Launch tags:Artificial Intelligence
Launch Team / Built With
Render: Get up to $100K in startup creditsCloud infrastructure for founders who want to move fast and NOT break things
Promoted
We built MemoryPlugin because we were tired of feeling productive while still wasting time re-teaching every model who I am, what I’m building, and where we left off. OpenClaw helped a ton, but my “memory” was still trapped per tool and per session. This is my attempt to make AI feel like one continuous workspace: connect once, keep your context consistent everywhere, and actually build momentum. I’d love your feedback on what should be remembered (and what absolutely shouldn’t).
okay this is exactly what i've been complaining about for months - why do i have to explain the same context to every single llm? love that you're syncing chrome and twitter bookmarks too, that's smart. question though: how are you handling the memory sync technically? is it through their apis or are you doing something more clever with mcps? i've had so many issues getting context to persist properly across sessions with our agents. also curious about privacy since you're basically storing everything, assuming it's all local? congrats and best of luck!"
@victor_eth Hey Victor, share the sentiment you expressed. It is frustrating to keep reiterating yourself.
On non-OpenClaw stuff: all sync is powered through the chrome-extension. It periodically also syncs non-desktop sessions (if you grant permission). For OpenClaw: The agent invokes our APIs on interactions.
On privacy: we operate on zero-trust, but available-everywhere principle. So we store on a cloud-vault for access and performance reasons. All LLM calls are made in saving-off mode. All data is encrypted at rest. And air-gapped architecturally. Happy to answer any more questions.
Can't wait for you to try and share your feedback.
Congrats on the launch! Treating memory as a shared workspace instead of something siloed per tool feels like a meaningful shift. How do you decide what information is promoted to long-term memory versus kept ephemeral, and what safeguards exist to prevent outdated or incorrect context from quietly persisting across sessions?
@vik_sh Great questions, Viktor.
1. How does information get promoted to long-term memory: We run a sophisticated process that emulates the human brain's sleep-process. No humans are involved in the process. LLMs are used on no-saving mode; along with a proprietary algorithm.
2. The same cycle also calibrates staleness of information based on triangulating timeline relevance from multiple data-points.
That's an impressive idea, it should definitely improve the workflow and save the time! Did i understand it correctly that this plugin is only for Chrome? Do you plan to expand the plugin to other browsers?
@ksenia_sh Thnaks Ksenia.
The OpenClaw plugin is agnostic of browser (integrates at API-level). The rest of the LLMs' context(Chatgpt, Claude, etc.) is currently supported through a Chrome-plugin (works on other Chromium-based browsers such as Brave as well). Which browser do you use?
@it_kabootar Understood, thanks for the answer! I'm using Brave
Does this store memories locally? It sounds like it requires storing memories in the cloud in maximem.ai ?
@jon_brown4 That's correct. Stores on cloud. Encrypted at rest. All LLM calls are in ZDR mode.
I'm unable to get an API key, it doesn't allow me to create one when I click it.
@billchirico Apologies for that Bill. Just to be sure, you are hitting app.maximem.ai --> Signing up for a free account --> Going to the API tab --> Generating key. Correct?
Would you mind sharing the specific issue / error you are facing? Would be glad to get on a call and help.
@rohit_waghire Hey Rohit, we maintain a zero-trust, encryption at rest, cloud-vault for the purpose. Only and only you can have access to it. (Not even us or our team).