# auto-terminal

> Use this tool when you need to automate terminal commands, manage background processes, and monitor output in real-time, with a GUI or MCP server interface that integrates with AI agents like Cursor and Claude Desktop. It solves problems of process management and terminal automation, accepting inputs like shell commands and program configurations, and outputting real-time logs and process status. Use it in contexts where automated process management and terminal interaction are required, such as development, testing, and deployment workflows.

Canonical page: https://skillsregistry.net/skills/hs03-auto-terminal  
JSON: https://api.skillsregistry.net/v1/skills/hs03-auto-terminal

## Description

# Auto Terminal

<img src="app_icon.png" width="128" />

[![Buy Me A Coffee](https://img.shields.io/badge/Buy%20Me%20A%20Coffee-FFDD00?style=for-the-badge&logo=buy-me-a-coffee&logoColor=black)](https://buymeacoffee.com/hs03)

**Auto Terminal** is a powerful process manager and terminal automation tool that doubles as a [Model Context Protocol (MCP)](https://modelcontextprotocol.io/) server. It allows AI agents (like Cursor, Claude Desktop) to execute terminal commands, manage background processes, and monitor output in real-time.

## Features

- **GUI & MCP Hybrid**: Runs as a standalone GUI application or a background MCP server.
- **Process Management**: Start, stop, and restart background processes easily.
- **Real-time Logs**: View live stdout/stderr logs for any running process.
- **Terminal Automation**: Execute arbitrary shell commands via MCP tools.
- **MacOS Optimized**: Native Dock icon support.

## Installation

```bash
git clone https://github.com/hoseoglee/auto-terminal.git
cd auto-terminal

# Create virtual environment
python3 -m venv .venv
source .venv/bin/activate

# Install dependencies
pip install -r requirements.txt
```

## Usage

### 1. GUI Mode
Run the application with a graphical interface to manage processes manually.

```bash
python gui_launcher.py
```

### 2. MCP Server Mode
This project uses `fastmcp` to expose tools to AI agents.

#### Arguments
- `--mcp`: Starts in MCP mode.
- `--sse`: Uses Server-Sent Events (SSE) transport (default on port 8000).

## MCP Configuration (for Cursor / Claude)

Add the following configuration to your `config.json` (Cursor) or `claude_desktop_config.json`.

**Option A: Run Directly (stdio)**
*Recommended for local usage.*
```json
{
  "mcpServers": {
    "auto-terminal": {
      "command": "/absolute/path/to/auto-terminal/.venv/bin/python",
      "args": [
        "/absolute/path/to/auto-terminal/gui_launcher.py",
        "--mcp"
      ]
    }
  }
}
```

**Option B: SSE Mode**
```json
{
  "mcpServers": {
    "auto-terminal": {
      "url": "http://localhost:8000/sse",
      "transport": "sse"
    }
  }
}
```

## Available Tools

The MCP server exposes the following tools to the AI:

- `list_programs()`: List configured and running processes.
- `start_program(name)`: Start a specific program.
- `stop_program(name)`: Stop a running program.
- `restart_program(name)`: Restart a program.
- `run_terminal_command(command, name, cwd)`: Execute any shell command.
- `get_program_logs(name, lines)`: Fetch recent logs.
- `send_program_input(name, input_text)`: Send stdin to a process.
- `add_program_config(name, command, cwd)`: Save a new program configuration.
- `remove_program_config(name)`: Remove a configuration.

## License

MIT License

## Trust

- **Trust score (0–1):** 0.50
- **Verification tier:** unverified

## Facts

- **Version:** 1.0.0
- **Skill type:** atomic
- **Execution layer:** container
- **Runtime environment:** vm
- **Category:** version-control
- **Updated:** 2026-05-14

## Source

- **Source listing:** [Smithery](https://smithery.ai/server/hs03/auto-terminal)

## Use it

Resolve this record through the SkillsRegistry MCP server (no auth, read-only):

```
claude mcp add --transport http --scope user skillsregistry https://api.skillsregistry.net/mcp
```

```json
{
  "jsonrpc": "2.0",
  "id": 1,
  "method": "tools/call",
  "params": {
    "name": "get_skill",
    "arguments": {
      "slug": "hs03-auto-terminal"
    }
  }
}
```

REST: `GET https://api.skillsregistry.net/v1/skills/hs03-auto-terminal` · pull for local use: `GET https://api.skillsregistry.net/v1/skills/hs03-auto-terminal/pull`

---
SkillsRegistry indexes agent skills from public registries and GitHub. Skills we have analysed are scanned with Circle-IR and scored on six dimensions; each listing states its scan coverage. More: https://skillsregistry.net/llms.txt