YoMemo.AI

Zero-trust, immutable memory for LLMs via MCP

2 followers

Zero-trust, immutable memory for LLMs via MCP

2 followers

YoMemo is a security-first memory relay for AI applications. Zero-trust encryption protects long-term and short-term memory at rest and in retrieval. Support for MCP, online key generation, and developer-friendly APIs.

Free Options

Launch tags:SaaS•Developer Tools•Tech

Launch Team

AWS Startup AdvisorPersonalized AWS guidance that meets you where you’re already building

Promoted

Maker

📌

Hi Product Hunt! 👋 I’m the solo developer behind YoMemo. I use AI assistants for everything—coding, brainstorming, and planning. But I constantly ran into the same wall: AI "amnesia." Every new session felt like starting from zero. I wanted my AI to remember my project architecture, my coding style, and my preferences, but I wasn't willing to compromise on privacy by storing that sensitive context in plain text on someone else's server. That’s why I built **YoMemo**. It’s designed to be a "Zero-Trust" memory layer for the AI era. Here’s why it’s different: 🔐 **True Zero-Trust:** Your data is encrypted client-side using RSA-OAEP + AES-GCM before it ever leaves your machine. I can’t read your memories, and neither can any server admin. You hold the keys. 🔌 **MCP Native:** By supporting the Model Context Protocol (MCP), YoMemo works seamlessly as a tool within Claude Desktop and other compatible hosts. Your AI can autonomously store and retrieve context when it needs it. 🚀 **LLM Agnostic:** Whether you use Claude, ChatGPT, or local models, your memory stays with you, not locked into a single provider. I’m officially launching today and would love to hear your thoughts! - How are you currently managing long-term context for your AI workflows? - What other integrations would make this even more useful for you? I’ll be here all day to answer your questions! Thanks for the support! ❤️

Report

10mo ago

Maker

Hello Product Hunt community! 👋

I’m the creator of YoMemo.ai, and today I’m thrilled to share a major architectural leap in how we handle personal knowledge.

The Problem: Most AI note-taking apps use vector search (RAG). It’s great for finding "similar" content, but it struggles with hard logic. It can't always connect the dots or guarantee 100% accuracy when you ask complex questions about your own life or data.

The Solution: Memory → Fact → Logic We’ve introduced a new engine that converts your unstructured Memories into structured Facts. By integrating Prolog—the gold standard for logic programming—YoMemo can now "reason" through your data rather than just "guessing" it.

How it works (The Example): Imagine you’ve jotted down these notes over months:

"Met Alice at the Web3 conference in Paris."
"Alice works for Protocol Labs."

Previously, an AI might just show you these notes. Now, YoMemo extracts:

attended(me, web3_conf, paris).
works_at(alice, protocol_labs).

The result? You can now ask: "Which of my contacts in the blockchain space have I met in Europe?" and get a mathematically certain answer.

Key Features:

✨ Automated Fact Extraction: Converts natural language to logic predicates.
🧠 Prolog Reasoning: Perform complex queries that standard AI search misses.
🔒 Privacy First: Your logical knowledge graph belongs to you.

I’d love to hear your thoughts! Does the bridge between LLMs and Symbolic Logic (Prolog) excite you as much as it does me?

I'll be here all day to answer your questions!

Check it out at: [yomemo.ai]