# Engram MCP server

> Provides persistent, project-aware memory for AI coding agents using a local SQLite database and full-text search.

- Repository: https://github.com/Gentleman-Programming/engram
- Page: https://hostedmcp.org/servers/gentleman-programming/engram
- Transport: stdio
- Stars: 8162
- License: MIT

## Install

### Claude Code

```sh
claude mcp add engram -- npx -y gentle-engram
```

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

```json
{
  "mcpServers": {
    "engram": {
      "command": "npx",
      "args": [
        "-y",
        "gentle-engram"
      ]
    }
  }
}
```

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

```json
{
  "servers": {
    "engram": {
      "type": "stdio",
      "command": "npx",
      "args": [
        "-y",
        "gentle-engram"
      ]
    }
  }
}
```

### VS Code CLI

```sh
code --add-mcp '{"name":"engram","type":"stdio","command":"npx","args":["-y","gentle-engram"]}'
```

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

```json
{
  "mcpServers": {
    "engram": {
      "command": "npx",
      "args": [
        "-y",
        "gentle-engram"
      ]
    }
  }
}
```

### Codex

```sh
codex mcp add engram -- npx -y gentle-engram
```

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

```toml
[mcp_servers.engram]
command = "npx"
args = ["-y", "gentle-engram"]
```

### Gemini CLI

```sh
gemini mcp add -s user engram npx -- -y gentle-engram
```

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

```json
{
  "mcpServers": {
    "engram": {
      "command": "npx",
      "args": [
        "-y",
        "gentle-engram"
      ]
    }
  }
}
```

## FAQ

### Which AI agents are compatible with Engram?

It works with many agents including Claude Code, OpenCode, Gemini CLI, Codex, VS Code (Copilot), Cursor, and Windsurf.

### Does Engram require external dependencies like Node.js or Docker?

No, it is a single Go binary that uses SQLite and requires no Node.js, Python, or Docker.

### Can I browse the memory stored by my agent?

Yes, you can use the built-in Terminal User Interface (TUI) to browse and search what your agent remembers.
