Product
M.N.E.M.O. is a platform for both end users and developers. It stores what you give it as exact, structured records, then lets you use that memory in three distinct ways: chat with it, share it with other people and organisations, or call it from your own RAG pipeline. They are three doors into the same memory, so you can start with one and add the others as you need them.
1 · AI chatbot
Upload documents and spreadsheets, then ask in plain language. Get the exact stored value back, with totals computed rather than guessed.
Plans: every plan (Free has a limited monthly allowance).
See how it works →2 · Knowledge Sharing Network
Share a subject privately with chosen people, or list it for anyone to subscribe to. Subscribers query it like their own, and you stay the owner.
Plans: subscribing on every plan; publishing and private sharing on Standard and above.
See how it works →3 · MnemoRAG
Define an endpoint over exactly the subjects you choose and call it from your app, bot or workflow. No session to manage.
Plans: Pro and Enterprise.
See how it works →Who it's for
You don't need to write code to get value from M.N.E.M.O., and you don't have to give up the interface when you do. The same memory serves both, so what an end user builds up in chat is exactly what a developer's application can query.
End users
Upload your documents and spreadsheets, chat with them, subscribe to knowledge other people share, and publish your own, all from the Dorothy app. A founder, analyst, support lead or consultant can have a working memory in minutes.
Developers
Standard and above can create an API token and talk to the memory directly: a REST API, an MCP server for Claude Desktop, a Skill for Claude Code, and, on Pro and Enterprise, MnemoRAG as a stateless RAG endpoint for your own pipeline.
Better together
The features are distinct, but they are not separate products. They all read the same subjects, so what one can see the others can see, and you can move from one to the next without re-uploading anything.
| AI chatbot | Knowledge Sharing Network | MnemoRAG | |
|---|---|---|---|
| Who asks | You and your team | Subscribers, in their own chat or app | Your application or workflow |
| How | Conversation in Dorothy, Claude Desktop or Claude Code | Publish or grant once, then they subscribe | One stateless POST call |
| What it reads | Your subjects and the ones shared with you | The subjects you choose to share | Exactly the subjects you put in the endpoint |
| Plans | All plans | Subscribe on all plans; publish on Standard and above | Pro and Enterprise |
A consultancy loads a client's data once. Its team chats with it to prepare the review. It shares a curated subject privately with the client, who can ask questions of it at any time. The client's own portal calls aMnemoRAG endpoint scoped to that shared subject. One memory, kept up to date in one place, and no file ever left it.
Most people begin with chat to get their own knowledge working, add sharing when someone else needs the answers, and add a MnemoRAG endpoint when a product or workflow needs to ask on its own. Each step reuses what is already in memory.
Proof of concept
Dorothy is Panaxon AI's own reference application, a conversational interface that shows what persistent, structured memory and knowledge sharing feel like day to day. It is one thing built on M.N.E.M.O., not a limit on what you can build.
Try it live
See the memory in action in a live chat app.
Next step
Choose a plan, or talk with Panaxon AI about your use case.Chat and subscribing are on every plan, publishing and private sharing start at Standard, and MnemoRAG is on Pro and Enterprise.
