ipIterPrompt
MCP serverKnowledge & MemoryVerified Last verified 2026-07-24

Knowledge Graph Memory MCP Server

by Model Context Protocol

A reference MCP server that stores information as a knowledge graph of entities, relations, and observations, so an agent can remember facts about you and your projects across conversations. The graph persists locally and the agent can read from and write to it, enabling continuity that a single chat context can't provide.

Works with

The Model Context Protocol (MCP).

When to use it

  • Giving an assistant durable memory of your preferences and projects
  • Maintaining context that should outlive a single chat session
  • Building a lightweight personal knowledge base the agent maintains

When not to

  • Storing sensitive data you don't want written to a local memory file

What it does

  • Store entities, relations, and observations as a graph
  • Recall remembered facts across separate conversations
  • Update and extend the graph as new information arrives

How to install

Setup steps with copy-paste config. Sources are cited under each method.

npx

Requires: Node.js

  1. Add to your MCP config

    The graph persists to a local file; no external service or key is needed.

    json
    {
      "mcpServers": {
        "memory": {
          "command": "npx",
          "args": ["-y", "@modelcontextprotocol/server-memory"]
        }
      }
    }
  2. Tell the agent what to remember

    Ask it to store key facts; on later sessions, ask it to recall them.

How to prompt it

Copy-paste prompts written to exercise this plugin, with what to expect.

Remember project context

You want the agent to retain facts about your stack across sessions.

Remember that this project uses Postgres, deploys on Fridays, and that Ana owns the billing service. Next time I ask, recall these.

What to expect: The agent writes those facts into the memory graph and can recall them in a later, separate conversation.

  • Be explicit about what to store; the agent only remembers what it records.

Tips for getting the most out of it

  • Review what gets written — the memory file is a plain local store you can inspect.
  • Combine with a system prompt that tells the agent when to consult memory.

Prompts to pair with it

Real prompts from the IterPrompt library you can copy or run.

Agent Orchestration Context Manager Agent

Elite AI context engineering specialist mastering dynamic context management, vector databases, knowledge graphs, and intelligent memory systems. Orchestrates context across multi-

3,888 copiesOpen ↗

MoltPass Client -- Cryptographic Passport for AI Agents

Turn any AI chat into a moltpass client -- cryptographic passport for ai agents with this community persona prompt.

3,877 copiesOpen ↗
4.6

Langchain Architecture

Design LLM applications using LangChain 1.x and LangGraph for agents, memory, and tool integration. Use when building LangChain applications, implementing AI agents, or creating co

2,284 copiesOpen ↗

Backend Development GraphQL Architect Agent

Master modern GraphQL with federation, performance optimization, and enterprise security. Build scalable schemas, implement advanced caching, and design real-time systems. Use PROA

747 copiesOpen ↗

Database Cloud Optimization Backend Architect Agent

Expert backend architect specializing in scalable API design, microservices architecture, and distributed systems. Masters REST/GraphQL/gRPC APIs, event-driven architectures, servi

4,775 copiesOpen ↗

DGX Spark Ops Engineer Agent

NVIDIA DGX Spark environment doctor for GB10/aarch64/CUDA-13 systems. Diagnoses and fixes ML stack setup, unified-memory, and thermal issues. Use PROACTIVELY when preparing or debu

5,091 copiesOpen ↗
See more prompts in the library

Frequently asked questions

Where is the memory stored?+

In a local file managed by the server, structured as a knowledge graph. Nothing is sent to a third-party service by the reference server.

Pairs well with

Alternatives to Knowledge Graph Memory MCP Server

Sources & verification

The facts on this page and where they come from. Anything unverified is labeled — we'd rather say "unknown" than guess.

Install steps, pricing, and availability change often. Always confirm on the official page before relying on setup details.