Omega Memory

v1.0.0Knowledge & Memorystable

Persistent memory for AI coding agents

ai-agentai-memoryclaudeclaude-codecoding-agent
Share:
145
Stars
0
Downloads
0
Weekly
0/5

What is Omega Memory?

Omega Memory is a Model Context Protocol (MCP) server that allows AI assistants like Claude, Cursor, and VS Code to persistent memory for ai coding agents

Persistent memory for AI coding agents

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

Features

  • Persistent memory for AI coding agents

Use Cases

Store persistent memory for coding agents
Enable cross-session learning and context
Maintain agent state between sessions
omega-memory

Maintainer

LicenseApache-2.0
Languagepython
Versionv1.0.0
UpdatedMay 21, 2026
Statushealthy
Maintenanceactive

Works with

ClaudeOpenAIwindowsmacoslinux

Installation

Manual Installation

npx omega-memory

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 Omega Memory

Omega Memory is a local-first persistent memory MCP server for AI coding agents that stores knowledge, decisions, lessons, and context in a local SQLite database with semantic search powered by an ONNX embedding model. It provides over 25 MCP tools covering memory storage and retrieval, session management, cross-session analytics, reminders, and knowledge-graph traversal, enabling agents to remember architectural decisions, past errors, and project preferences across separate conversations. The system runs entirely on-device with no external API calls required, making it suitable for private or offline development environments.

Prerequisites

  • Python 3.9 or later with pip available
  • An MCP-compatible client: Claude Desktop, Claude Code, Cursor, Windsurf, or Cline
  • Approximately 100 MB of disk space for the ONNX embedding model at ~/.cache/omega/models/
  • No external API keys — all processing is local
1

Install omega-memory with server support

Install the package with the server extra to include the MCP server entrypoint.

pip install omega-memory[server]
2

Run the setup wizard for your client

The setup wizard initialises the local database, downloads the embedding model, and writes the MCP configuration for your chosen client.

# For Claude Desktop
omega setup --client claude-desktop

# For Cursor
omega setup --client cursor

# For Claude Code
omega setup --client claude-code
3

Verify the setup is healthy

Run the built-in doctor command to confirm the database, embedding model, and client registration are all operational.

omega doctor
4

Restart your MCP client

Fully quit and relaunch your client (Claude Desktop, Cursor, etc.). Omega's memory tools — omega_store, omega_query, omega_welcome, omega_lessons, and others — will appear in the tools panel.

5

Store your first memory

In a conversation, ask the AI to store a memory. Omega will persist it and surface it automatically in future sessions when relevant context is detected.

Omega Memory Examples

Client configuration

Manual claude_desktop_config.json entry for Omega Memory if you prefer not to use the setup wizard.

{
  "mcpServers": {
    "omega-memory": {
      "command": "python",
      "args": ["-m", "omega.server"]
    }
  }
}

Prompts to try

Example prompts for storing, retrieving, and managing persistent memories across coding sessions.

- "Store a decision: we use PostgreSQL for this project, not MySQL — add it under the 'architecture' type"
- "What do you remember about the authentication implementation in this project?"
- "Save a lesson learned: always add database indexes before running load tests in staging"
- "Show me all the errors I encountered last week across my coding sessions"
- "Give me a weekly digest of what I've built and what decisions I've made this week"

Troubleshooting Omega Memory

omega setup fails with model download errors

The ONNX embedding model is downloaded from HuggingFace on first setup. Check your internet connection, then run omega setup again. If behind a proxy, set HTTP_PROXY and HTTPS_PROXY environment variables before running the command.

Memory tools do not appear in Claude Desktop

Fully quit Claude Desktop (not just close the window) and relaunch it. Verify that omega setup --client claude-desktop completed without errors and that the MCP entry exists in your Claude Desktop settings.

Semantic search returns irrelevant memories

Run omega_consolidate to merge fragmented memories and improve the embedding index. Also check omega_type_stats to see if memories are correctly categorised — miscategorised memories can affect retrieval relevance.

Frequently Asked Questions about Omega Memory

What is Omega Memory?

Omega Memory is a Model Context Protocol (MCP) server that persistent memory for ai coding agents It connects AI assistants to external tools and data sources through a standardized interface.

How do I install Omega Memory?

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

Which AI clients work with Omega Memory?

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

Is Omega Memory free to use?

Yes, Omega Memory is open source and available under the Apache-2.0 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": { "omega-memory": { "command": "npx", "args": ["-y", "omega-memory"] } } }

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

Read the full setup guide →

Ready to use Omega Memory?

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