Memorizer

v1.0.0Knowledge & Memorystable

Vector-search powered agent memory MCP server

memorizermcpai-integration
Share:
164
Stars
0
Downloads
0
Weekly
0/5

What is Memorizer?

Memorizer is a Model Context Protocol (MCP) server that allows AI assistants like Claude, Cursor, and VS Code to vector-search powered agent memory mcp server

Vector-search powered agent memory MCP server

This server falls under the Knowledge & Memory category on MCPgee, the world's largest MCP server directory with 33,000+ servers.

Features

  • Vector-search powered agent memory MCP server

Use Cases

Vector-search powered agent memory MCP server
petabridge

Maintainer

LicenseMIT
Languagec#
Versionv1.0.0
UpdatedMay 4, 2026
Statushealthy
Maintenanceactive

Works with

ClaudeOpenAIwindowsmacoslinux

Installation

Manual Installation

npx memorizer

Configuration

Configuration Details

Config File

claude_desktop_config.json

Performance

Response Metrics

Response Time< 200ms
ThroughputMedium

Resource Usage

Memory UsageLow
CPU UsageLow

How to Set Up and Use Memorizer

Memorizer is a .NET-based vector-search powered memory service that allows AI agents to store, retrieve, and semantically search through memories using PostgreSQL with the pgvector extension. It provides persistent, versioned memory with workspace and project organization, relationship graphs between memories, and a web UI for manual management. Developers use it to give AI agents durable long-term memory that survives across sessions and can be queried by semantic similarity.

Prerequisites

  • Docker and Docker Compose installed on your system
  • An MCP-compatible client such as Claude Desktop, Cursor, or VS Code
  • Optionally an OpenAI API key or a locally running Ollama instance for embeddings
  • Port 5000 available for the Memorizer API and Web UI
1

Clone the repository

Clone the Memorizer repository from GitHub to get the Docker Compose files needed to run the full stack.

git clone https://github.com/petabridge/memorizer.git
cd memorizer
2

Start the full stack with Docker Compose

Run docker-compose to start Memorizer, PostgreSQL with pgvector, PgAdmin, and Ollama all at once. The Memorizer Web UI will be available at http://localhost:5000.

docker-compose up -d
3

Configure your embedding provider

Open the Memorizer Web UI at http://localhost:5000 and navigate to Provider Settings. Configure either Ollama (default, runs locally on port 11434) or OpenAI with your API key for generating vector embeddings.

4

Add the MCP server to your client configuration

Connect your MCP client to the Memorizer server using the Streamable HTTP transport. The MCP endpoint is at /mcp (note: changed from / in version 2.0).

{
  "mcpServers": {
    "memorizer": {
      "url": "http://localhost:5000/mcp"
    }
  }
}
5

Add a system prompt to your AI agent

For best results, add a system prompt or AGENT.md instruction telling the LLM to use the memorizer MCP tools (store, retrieve, search, list_memories) for persistent memory management across sessions.

Memorizer Examples

Client configuration

Configure Claude Desktop or another MCP client to connect to Memorizer using the Streamable HTTP transport at the /mcp endpoint.

{
  "mcpServers": {
    "memorizer": {
      "url": "http://localhost:5000/mcp"
    }
  }
}

Prompts to try

Example prompts that demonstrate Memorizer's persistent memory and semantic search capabilities.

- "Remember that the production database password is stored in 1Password under 'prod-db-v3'"
- "Search my memories for anything related to the authentication architecture decisions"
- "What do you remember about our discussion on microservices last week?"
- "Create a memory tagged 'architecture' about the decision to use event sourcing for the orders service"
- "Show me all memories in the 'Project Alpha' workspace"

Troubleshooting Memorizer

MCP client cannot connect after upgrading from Memorizer 1.x

In version 2.0, the MCP endpoint moved from http://localhost:5000 to http://localhost:5000/mcp. Update your client configuration URL accordingly.

Embeddings are not being generated or search returns no results

Check that your embedding provider is correctly configured in the Web UI under Provider Settings. If using Ollama, ensure the Ollama container is running and the selected embedding model has been pulled (e.g., run 'docker exec <ollama-container> ollama pull nomic-embed-text').

docker-compose up fails with port conflict errors

Memorizer uses ports 5000 (API/UI), 5432 (PostgreSQL), 5050 (PgAdmin), and 11434 (Ollama). Stop any services already using those ports or edit docker-compose.yml to remap them before starting.

Frequently Asked Questions about Memorizer

What is Memorizer?

Memorizer is a Model Context Protocol (MCP) server that vector-search powered agent memory mcp server It connects AI assistants to external tools and data sources through a standardized interface.

How do I install Memorizer?

Follow the installation instructions on the Memorizer GitHub repository. Clone the repo, install dependencies, and add the server config to your AI client.

Which AI clients work with Memorizer?

Memorizer works with all major MCP-compatible AI clients including Claude Desktop, Claude Code, Cursor, VS Code (GitHub Copilot), Windsurf, and Cline.

Is Memorizer free to use?

Yes, Memorizer is open source and available under the MIT license. You can use it freely in both personal and commercial projects.

Browse More Knowledge & Memory MCP Servers

Explore all knowledge & memory servers available in the MCPgee directory. Each server includes setup guides for Claude, Cursor, and VS Code.

Quick Config Preview

{ "mcpServers": { "memorizer": { "command": "npx", "args": ["-y", "memorizer"] } } }

Add this to your claude_desktop_config.json or .cursor/mcp.json

Read the full setup guide →

Ready to use Memorizer?

Browse our complete directory of 33,000+ MCP servers, read setup guides for your editor, and start building with the Model Context Protocol.

33,000+ ServersFree & Open SourceStep-by-Step Guides