doobidoo/mcp-memory-service

Open-source persistent memory for AI agent pipelines (LangGraph, CrewAI, AutoGen) and Claude. REST API + knowledge graph + autonomous consolidation.

What it solves

AI agents typically suffer from "amnesia," losing all context and project history once a session ends. This project provides a persistent, shared memory backend that allows agents to store decisions, code patterns, and architecture context, ensuring that new sessions start with existing knowledge rather than from zero.

How it works

It operates as a self-hosted service that can be accessed via a REST API or the Model Context Protocol (MCP). It uses AI embeddings (run locally via ONNX) for semantic search and supports multiple storage backends including SQLite, Milvus, and Cloudflare. The system organizes information as a knowledge graph with typed edges to track causal relationships between facts, rather than just storing isolated snippets of text.

Who it’s for

Developers building agentic pipelines with frameworks like LangGraph, CrewAI, or AutoGen, as well as users of MCP-compatible clients like Claude Desktop, Cursor, and Windsurf who want their AI assistants to remember details across different chats and tools.

Highlights

  • Multi-Transport Support: Accessible via REST API, MCP, and a web dashboard.
  • Knowledge Graph: Visualizes and manages memories as an interactive 3D galaxy of nodes and relationships.
  • Local-First Privacy: Embeddings run locally via ONNX to keep data within your own infrastructure.
  • Framework Agnostic: Works with any HTTP client and integrates with major AI IDEs and desktop apps.
  • Flexible Backends: Supports local SQLite for speed, Milvus for dedicated vector search, and Cloudflare for team synchronization.

Related

  • Project
    Lyellr88/marm-memory

    A persistent local memory server for AI agents using the Model Context Protocol (MCP), providing shared long-term memory and code-graph awareness across different AI clients.

  • Project
    MaxFreedomPollard/Compartment

    An encrypted, fully offline persistent memory vault for AI agents that allows them to share learned facts and preferences across sessions without any network access or telemetry.

  • Project
    peterfei/ai-agent-team

    A virtual AI development team with specialized roles (PM, Backend, Frontend, etc.) and a persistent memory system for managing complex software projects.

  • Project
    rahilp/second-brain-cloudflare

    Second Brain is a Cloudflare‑hosted personal + team memory system that stores notes, decisions, and project facts and makes them searchable by meaning. It integrates with Claude, ChatGPT, Cursor, Obsidian, Notion, calendars, etc., and provides a secure, self‑hosted backend (D1, Vectorize, Workers AI) plus a “Prompt Capsule” feature for deterministic LLM context.

  • Project
    mnemon-dev/mnemon

    Mnemon is a cross‑platform CLI that adds persistent, graph‑based memory to LLM agents. The host LLM supervises what to remember, link, or forget, while Mnemon handles deterministic storage, decay, deduplication, and optional vector search. It integrates with many AI coding tools (Claude Code, Codex, Cursor, etc.) via one‑click setup, requiring only a single local binary and no extra API keys.