# Memory Engine MCP server

> Provides AI assistants with local-first graph memory using SQLite and semantic search to recall, connect, and learn from long-term context.

- Repository: https://github.com/SimoneB79/memory-engine-mcp
- Page: https://hostedmcp.org/servers/simoneb79/memory-engine-mcp
- Transport: remote
- Stars: 1
- License: MIT
- Latest version: 1.5.3

## Install

### Claude Code

```sh
claude mcp add memory-engine-mcp -- docker run -i --rm ghcr.io/simoneb79/memory-engine-mcp:1.5.3
```

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

```json
{
  "mcpServers": {
    "memory-engine-mcp": {
      "command": "docker",
      "args": [
        "run",
        "-i",
        "--rm",
        "ghcr.io/simoneb79/memory-engine-mcp:1.5.3"
      ]
    }
  }
}
```

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

```json
{
  "servers": {
    "memory-engine-mcp": {
      "type": "stdio",
      "command": "docker",
      "args": [
        "run",
        "-i",
        "--rm",
        "ghcr.io/simoneb79/memory-engine-mcp:1.5.3"
      ]
    }
  }
}
```

### VS Code CLI

```sh
code --add-mcp '{"name":"memory-engine-mcp","type":"stdio","command":"docker","args":["run","-i","--rm","ghcr.io/simoneb79/memory-engine-mcp:1.5.3"]}'
```

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

```json
{
  "mcpServers": {
    "memory-engine-mcp": {
      "command": "docker",
      "args": [
        "run",
        "-i",
        "--rm",
        "ghcr.io/simoneb79/memory-engine-mcp:1.5.3"
      ]
    }
  }
}
```

### Codex

```sh
codex mcp add memory-engine-mcp -- docker run -i --rm ghcr.io/simoneb79/memory-engine-mcp:1.5.3
```

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

```toml
[mcp_servers.memory-engine-mcp]
command = "docker"
args = ["run", "-i", "--rm", "ghcr.io/simoneb79/memory-engine-mcp:1.5.3"]
```

### Gemini CLI

```sh
gemini mcp add -s user memory-engine-mcp docker -- run -i --rm ghcr.io/simoneb79/memory-engine-mcp:1.5.3
```

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

```json
{
  "mcpServers": {
    "memory-engine-mcp": {
      "command": "docker",
      "args": [
        "run",
        "-i",
        "--rm",
        "ghcr.io/simoneb79/memory-engine-mcp:1.5.3"
      ]
    }
  }
}
```

## Environment variables

| Name | Required | Secret | Description |
|---|---|---|---|
| `MARKDOWN_SOURCE` | no | no | Optional directory for one-way markdown import. |
| `MEMORY_API_TOKEN` | no | no |  |
| `MEMORY_DB_PATH` | no | no | SQLite database path inside the container. |
| `MEMORY_PORT` | no | no | SSE server port. Default: 8085. |

## Tools (36)

- `remember`
- `recall`
- `working_set`
- `semantic_search`
- `get_atom`
- `list_atoms`
- `merge_atoms`
- `export_atom`
- `search_graph`
- `suggest_bonds`
- `suggest_bonds_all`
- `curator_run`
- `cognitive_status`
- `learning_run`
- `decay_run`
- `cleanup_sessions`
- `cleanup_duplicates`
- `reindex_embeddings`
- `error_check`
- `error_log`
- `error_list`
- `preference_search`
- `import_markdown`
- `memory_summary`
- `stats`
- `version`
- `recall_session`
- `session_summary`
- `memory_contradict`
- `list_contradictions`
- `classify_memory_tier`
- `memory_impact`
- `backup_database`
- `restore_database`
- `export_all`
- `import_data`

## FAQ

### Does Memory Engine require a cloud API?

No, it is a local-first system using a SQLite database and optional local embeddings via Ollama, requiring no cloud API.

### Can I use my existing Markdown notes?

Yes, you can import existing notes one-way into the memory system without replacing your human-readable files.

### Is there a way to visualize the memory graph?

Yes, an optional web UI is available for graph exploration, atom inspection, and impact analysis.