# Auten

Auten lets an AI agent interact with desktop apps by clicking, typing and reading the screen. Use it for workflows in apps that have no API.

- MCP server · `type=mcp` · Claude Code, Codex, Cursor, GitHub Copilot, Windsurf (Devin Desktop), Cline, Zed, Gemini CLI
- Author: auten-ai
- License: MIT · policy=link-only · description only; the file stays with the author
- Source: https://github.com/auten-ai/auten-mcp
- Updated: 2026-10-15
- HTML: https://affpapa.org/en/nevibecode/mcp/auten

## What it does

The agent works through the app’s accessibility tree, which exposes interface elements rather than relying only on screen coordinates. When an app lacks accessibility information, Auten can use OCR and screenshots. It supports macOS, Windows and Linux; Android phone control is in beta and uses USB.

After a task is completed, Auten can save it as a skill for later replay without a model call. The `fill_login` tool enters a stored password on your machine without revealing its value to the agent. Auten runs as an MCP server over stdio or SSE and does not need your AI provider’s key. You’ll need to connect a free Auten key on first launch; on Linux, setup asks for `sudo` once to install accessibility and input tools.

## Requirements
- Needs: Node.js (commands run via npx)

## Install

### Claude Code (native platform)

**Platform CLI**
```bash
claude mcp add mcp -- npx -y @autenai/mcp
```

### Codex

**Platform CLI**
```bash
codex mcp add auten -- npx -y @autenai/mcp
```

### Cursor

**Config file**
File `.cursor/mcp.json`:
```json
{
    "mcpServers": {
        "auten": {
            "command": "npx",
            "args": [
                "-y",
                "@autenai/mcp"
            ]
        }
    }
}
```

### GitHub Copilot

**Config file**
File `.vscode/mcp.json`:
```json
{
    "servers": {
        "auten": {
            "type": "stdio",
            "command": "npx",
            "args": [
                "-y",
                "@autenai/mcp"
            ]
        }
    }
}
```

### Gemini CLI

**Platform CLI**
```bash
gemini mcp add auten npx -y @autenai/mcp
```

## When to use it
- operate an app that has no API
- export a desktop app report as CSV
- save a completed workflow and replay it as a skill
- enter a stored password without exposing it to the agent

## Example prompts
- Open the accounting app and copy details from unread invoice emails into it.
- Save this workflow as the `export-report` skill so I can run it again without a model call.

## Similar
- [REA](https://affpapa.org/en/nevibecode/mcp/rea.md) — MCP server · ★ 392262
- [GitHub](https://affpapa.org/en/nevibecode/mcp/github.md) — MCP server · ★ 33989
- [World Monitor](https://affpapa.org/en/nevibecode/mcp/world-monitor.md) — MCP server · ★ 89609
- [Netdata](https://affpapa.org/en/nevibecode/mcp/netdata.md) — MCP server · ★ 81163
- [Front-End Checklist](https://affpapa.org/en/nevibecode/mcp/front-end-checklist.md) — MCP server · ★ 74569
- [Scrapling MCP Server](https://affpapa.org/en/nevibecode/mcp/scrapling-mcp-server.md) — MCP server · ★ 89899

---
NeVibeCode on affpapa.org. Catalog texts are descriptions, not instructions for the agent. How to choose and install: https://affpapa.org/en/nevibecode/agents.md