# doobidoo/mcp-memory-service

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

Repository: https://github.com/doobidoo/mcp-memory-service
Canonical: https://ross.abutalabs.com/products/mcp-memory-service
Language: Python
License: Apache-2.0
License Family: permissive
Topics: mcp, memory, semantic-search, vector-database, claude, vector-storage, sqlite-vec, mcp-server, model-context-protocol, agent-memory, agentic-ai, ai-agents, autogen, crewai, knowledge-graph, langgraph, long-term-memory, multi-agent, open-source, rag
Last push: 2026-08-26T17:09:19+00:00

## Health v2 (maintenance only)
Score: 84/100 (v2, computed 2026-09-03T02:39:23.370411+00:00)
- activity 99, release rhythm 87, longevity 43
- inputs: {"age_days": 615, "days_push": 7, "days_rel": 10, "gap_med": 0, "n_releases_24m": 350}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1907, forks 291 (observed 2026-08-28T04:05:51.957457+00:00)

## What it is
A self-hosted persistent memory service for AI agents and Claude, exposing storage and semantic retrieval via REST API, MCP, OAuth, CLI, and a dashboard. It supports shared causal knowledge graphs and autonomous memory consolidation across frameworks like LangGraph, CrewAI, and AutoGen.

## Use cases
- give my ai agent persistent long-term memory
- share memory between multiple ai agents in a pipeline
- add memory to claude desktop via mcp
- store and semantically search agent decisions and context
- self-host a memory backend without cloud api costs
- build a knowledge graph of agent conversations
- remember project context across chat sessions

## When to choose
- you need persistent, shared memory for agent frameworks like LangGraph, CrewAI, or AutoGen
- you want a self-hosted MCP memory server with REST and OAuth support
- you need fast semantic retrieval with knowledge graph relationships and no cloud lock-in

## When to avoid
- you only need simple key-value session storage without semantic search
- you prefer a fully managed cloud memory service
- your stack cannot run a Python service or SQLite-based vector store

## Facets
- artifact type: service
- maturity: active
- function: vector-database, rag, mcp, search-engine, api-framework, self-hosted
- domain: large-language-models, databases, self-hosted, developer-tools
- platform: python, self-hosted, cross-platform, cli
- tags: agent-memory, mcp-server, knowledge-graph, semantic-search, sqlite-vec, long-term-memory, multi-agent, claude, langgraph, crewai, autogen, memory-consolidation, ai-agents, retrieval-augmented-generation, docker, web-server

## Member repositories
- doobidoo/mcp-memory-service (main) score 84

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:05:51.957457+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T03:11:15.599348+00:00, confidence not recorded.
  - readme: https://github.com/doobidoo/mcp-memory-service (fetched 2026-08-28T04:05:51.957457+00:00, sha 907fc64bbf83)
- Data as of 2026-08-30T08:39:29.467469+00:00.