RichmondAlake/memorizz

MemoRizz: A Python library serving as a memory layer for AI applications. Leverages popular databases and storage solutions to optimize memory usage. Provides utility classes and methods for efficient data management.

What it solves

Memorizz is a framework for building AI agents that can remember information across sessions and conversations. It addresses the problem of "forgetting" in LLMs by providing a structured, multi-layered memory system and a control plane for continual learning, allowing agents to retain and refine knowledge over time.

How it works

It implements several memory types (episodic, semantic, procedural, short-term, and shared) and supports pluggable storage providers like MongoDB, Oracle, or the local filesystem. The framework uses semantic retrieval (embeddings and vector search) and entity memory for profile-style facts. It also features a "Meta-Harness" that can run other coding agents (like Claude Code or OpenHands) behind a shared memory and policy control plane.

Who it’s for

Developers building memory-augmented AI agents, researchers exploring continual learning in LLMs, and users who want a persistent, personalized AI assistant with a local web UI and CLI.

Highlights

  • Multi-layered Memory: Supports episodic, semantic, procedural, and short-term memory systems.
  • Continual Learning: Automatically promotes successful tool workflows into reusable skills.
  • Meta-Harness: A control plane to run external agents (e.g., Codex, OpenHands) with shared memory and host-verified outcomes.
  • MCP Connectivity: First-class support for Model Context Protocol (MCP) for connecting to tools like Notion and Google Calendar.
  • Local LLM Support: Native integration with Hugging Face, MLX (Apple Silicon), Ollama, and OpenAI-compatible servers.
  • Production Governance: Includes durable host approvals, sandboxed code execution, and context-window telemetry.

Related

  • Project
    sopaco/cortex-mem

    A Rust-based long-term memory framework for AI agents that uses a three-tier hierarchical architecture and a virtual filesystem to provide persistent, context-aware interactions.

  • Project
    MemoriLabs/Memori

    A persistent, structured memory layer for AI agents and LLMs that automatically captures and recalls user preferences and agent actions to reduce token costs and maintain context across sessions.

  • Project
    mem9-ai/mem9

    A persistent memory layer for AI agents that enables shared context and recall across sessions, machines, and different agent runtimes.

  • Project
    TencentCloud/TencentDB-Agent-Memory

    A shared memory server and management hub that allows AI agents to persist and share experience, skills, and project knowledge across different sessions and frameworks.

  • Dispatch
    Exploring Memory Systems for AI Agents

    An exploration of the current landscape of memory systems for AI agents, focusing on thetextual analysis of how developers are approaching thetextual analysis of state management and context window managements