a2328275243/mempalace-evolve

Self-evolving memory palace for AI agents — persistent memory with automatic learning, knowledge graph, and multi-agent support

What it solves

MemPalace Evolve provides a local-first, long-term memory layer for AI agents. It solves the problem of agents losing context over time by allowing them to store durable facts, track relationships, and isolate data for different users, projects, or agents into separate boundaries.

How it works

It functions as a memory component that can be integrated via a Python SDK, an MCP (Model Context Protocol) client, or a REST API. The system organizes information using a hierarchy: "wings" act as hard boundaries for tenants or projects, and "rooms" act as categories within those wings. It supports storing memories with metadata and tags, adding factual relationships, and uses an evolve() function to score, promote, and clean stored context over time.

Who it’s for

Developers building AI agents or applications that require persistent, structured memory across sessions, as well as users of MCP-capable clients like Claude Code or Cursor.

Highlights

  • Multi-tenant Isolation: Uses "wings" to keep different projects or users completely separate.
  • Flexible Integration: Accessible through a Python SDK, REST API, or MCP server.
  • Knowledge Graph Capabilities: Supports adding and querying facts to track relationships between entities.
  • Lifecycle Management: Includes an evolve process to maintain memory quality and a doctor tool for health checks.

Related

  • Project
    MemPalace/mempalace

    A local-first AI memory system that stores conversation history verbatim and uses semantic search for high-accuracy retrieval without needing cloud APIs.

  • Project
    agentic-box/memora

    An MCP memory layer for AI agents that provides persistent, structured storage and semantic retrieval of facts, tasks, and documents via a knowledge graph.

  • Project
    MemMachine/MemMachine

    An open-source long-term memory layer for AI agents that enables persistent storage and recall of user preferences and conversational context across sessions.

  • Project
    MemTensor/memmy-agent

    A personal memory infrastructure that allows different AI agents to share a structured memory of user preferences and project history, ensuring task continuity across tools.

  • Project
    MaxFreedomPollard/Compartment

    Compartment is an encrypted, offline vault that lets AI agents store, retrieve, and manage session memories locally, providing fast hybrid search and privacy‑preserving persistence without any network or API dependencies.