matrixorigin/memoria

Secure memory management for AI Agents • Ensures data integrity • Reduces hallucinations • Maintains consistent long-term context

What it solves

Memoria provides a persistent memory layer for AI agents that allows them to remember facts, preferences, and decisions across different sessions. It solves the problem of memory instability and lack of control by introducing Git-like version control for agent memory, enabling users and agents to track, audit, and reverse changes to the agent's knowledge base.

How it works

Memoria acts as a bridge between AI agents and a storage backend powered by MatrixOne's Copy-on-Write engine. It uses a Model Context Protocol (MCP) server to provide agents with a set of tools for storing, retrieving, and managing memory. The system supports hybrid retrieval (combining vector and full-text search) and implements versioning features such as snapshots, branching, and merging to isolate experiments or roll back errors.

Who it’s for

Developers building AI agents (using tools like Cursor, Claude Code, or Kiro) who need their agents to have long-term, auditable memory that persists across conversations.

Highlights

  • Git-for-Memory: Supports zero-copy branching, instant snapshots, and point-in-time rollbacks for memory mutations.
  • Hybrid Search: Combines vector and full-text search to find memories by meaning and keywords.
  • Self-Governance: Automatically detects contradictions and quarantines low-confidence memories.
  • Flexible Deployment: Available as a cloud service or self-hosted via Docker for full data control.
  • Steering Rules: Provides guidance for agents on when to store, retrieve, or purge memory based on the conversation lifecycle.

Related

  • Project
    zhangfengcdt/memoir

    A high-performance semantic memory system for AI agents that introduces Git-like version control, hierarchical semantic paths, and cryptographic integrity to memory management.

  • Project
    AgentMemoryRepo/agentmemoryrepo

    A Git-based memory standard that allows AI agents to maintain persistent, linked knowledge across sessions, users, and agent swarms.

  • Dispatch
    Why Versioned Markdown Folders are the Ideal 'Brain' for AI Agents

    Explore why a simple architecture of Git-backed markdown files is replacing complex vector databases as the primary memory system for production AI agents.

  • Project
    atomicstrata/atomicmemory

    AtomicMemory is an open‑source, inspection‑ready memory service for LLM agents. It offers correction‑aware updates, model‑agnostic embeddings, and both managed‑cloud and self‑hosted Docker deployments. The project ships a Rust Core server, a TypeScript SDK, a CLI (`am`), adapters for LangChain, LangGraph, Vercel AI, etc., and plugins for various agent hosts. Quick start is a one‑line installer that creates a cloud profile; a `--local` flag runs the Core container on your machine. Memory can be ingested and queried via the CLI or programmatically via the SDK, and agents can be linked through an MCP server. Licensed under Apache 2.0.

  • Project
    mcncarl/agent-memory-vault

    A Python‑based, local‑first system that lets Claude Code, Codex and similar agents store and retrieve shared knowledge in a private Markdown vault. The vault is versioned with Git, indexed with SQLite (and optionally a local embedding vector store), and all writes go through a strict read‑prepare‑authorize‑apply workflow that records immutable receipts. It provides searchable, auditable long‑term memory for AI coding assistants while keeping the data private to the user.