# OMEGA MCP server

> Provides persistent, local-first semantic memory for AI coding agents to retain decisions, preferences, and context across sessions.

- Repository: https://github.com/omega-memory/core
- Page: https://hostedmcp.org/servers/omega-memory/core
- Transport: stdio
- Stars: 226
- License: Apache-2.0
- Latest version: 0.9.3

## Install

### Claude Code

```sh
claude mcp add core -- uvx omega-memory
```

### Cursor (`~/.cursor/mcp.json`)

```json
{
  "mcpServers": {
    "core": {
      "command": "uvx",
      "args": [
        "omega-memory"
      ]
    }
  }
}
```

### VS Code (`.vscode/mcp.json`)

```json
{
  "servers": {
    "core": {
      "type": "stdio",
      "command": "uvx",
      "args": [
        "omega-memory"
      ]
    }
  }
}
```

### VS Code CLI

```sh
code --add-mcp '{"name":"core","type":"stdio","command":"uvx","args":["omega-memory"]}'
```

### Claude Desktop (`claude_desktop_config.json`)

```json
{
  "mcpServers": {
    "core": {
      "command": "uvx",
      "args": [
        "omega-memory"
      ]
    }
  }
}
```

### Codex

```sh
codex mcp add core -- uvx omega-memory
```

### Codex config.toml (`~/.codex/config.toml`)

```toml
[mcp_servers.core]
command = "uvx"
args = ["omega-memory"]
```

### Gemini CLI

```sh
gemini mcp add -s user core uvx -- omega-memory
```

### Gemini settings.json (`~/.gemini/settings.json`)

```json
{
  "mcpServers": {
    "core": {
      "command": "uvx",
      "args": [
        "omega-memory"
      ]
    }
  }
}
```

## Tools (14)

- `omega_store`
- `omega_query`
- `omega_welcome`
- `omega_protocol`
- `omega_checkpoint`
- `omega_resume_task`
- `omega_memory`
- `omega_profile`
- `omega_remind`
- `omega_maintain`
- `omega_stats`
- `omega_reflect`
- `omega_review`
- `context_packet`

## FAQ

### Does OMEGA work with different AI models?

Yes, it is designed to work with any major LLM and coding agent, including Claude, GPT, Gemini, Cursor, and Claw Code.
