A real Mac for your coding agent: build, test, run, and tap through iOS apps from any machine
- ✓Open-source license (MIT)
- ✓Actively maintained (<30d)
- ✓Clear description
- ✓Topics declared
- ✓Documented (README)
- !Install pipes a remote script into a shell (curl | sh)
git clone https://github.com/opslane/macbox-agent{
"mcpServers": {
"macbox-agent": {
"command": "node",
"args": ["/path/to/macbox-agent/dist/index.js"]
}
}
}MCP Servers overview
<p align="center"><img src="logo.png" width="96" alt="macbox"></p>
# macbox
**A real Mac for your coding agent.** macbox lets Claude Code, Codex, Cursor, or any MCP
client build, test, run, and tap through native iOS apps on a real Mac in the cloud, from
any machine, including Linux boxes and cloud agent sandboxes.
- **Build and test:** `xcodebuild` and your XCTest and UI tests, on a Mac mini running the
latest Xcode. Your agent gets the results and the error lines.
- **Run and drive the app:** launch it on the iOS simulator, read what's on screen, tap,
type, and swipe, so your agent can check its own work.
- **Watch live:** every running app gets a browser link where you can watch and use it too.
Pay per Mac minute, whole minutes only: $0.05, and a job under a minute is free. Get access
and free credit at **[macbox.build](https://macbox.build)**.
## Set up
The quickest way: tell your agent
> Set up macbox by following https://api.macbox.build/guide/
Or pick your client:
**Claude Code (plugin: skill + MCP tools)**
```sh
claude plugin marketplace add opslane/macbox-agent
claude plugin install macbox@macbox
```
**Any MCP client** (Codex, Cursor, Claude Desktop, VS Code, ...), with Node 18+:
```json
{
"mcpServers": {
"macbox": { "command": "npx", "args": ["-y", "macbox", "mcp"] }
}
}
```
**The CLI** (macOS or Linux):
```sh
curl -fsSL https://api.macbox.build/install.sh | sh
macbox login
macbox setup # installs the skill for Claude Code and Codex, and notes in CLAUDE.md / AGENTS.md
```
Log in with `macbox login`, or set `MACBOX_API_KEY` (in a cloud agent, as an environment
secret).
## What your agent gets
| MCP tool | CLI | What it does |
|---|---|---|
| `build` | `macbox build` | Build for the iOS simulator (or macOS, for a Mac app) |
| `test` | `macbox test` | Run tests on a fresh simulator; one PASS/FAIL line per test |
| `screenshot` | `macbox screenshot` | Launch the last build and return a picture |
| `run` | `macbox run` | Launch the app and keep it running; returns a live link |
| `describe`, `tap`, `type`, `swipe`, `press_button`, `look` | `macbox ui ...` | Use the running app like a person |
| `stop` | `macbox stop` | End the running app |
| `report_problem` | `macbox feedback` | Tell the macbox team when something on our side is wrong |
For deeper work there is also `macbox mcp --xcode`: Sentry's open-source
[MobileBuildMCP](https://www.mobilebuildmcp.com/) Xcode tools (LLDB debugging, Swift packages,
simulator settings), running on the macbox Mac.
## How it works
`macbox` sends the files git tracks (never `.env` files unless you ask) to one Mac mini,
into your own macOS VM that only you use and that can't reach other VMs or our network.
Only changed files are sent after the first time, and the build cache stays, so the second
build is fast. Native iOS apps work today: Xcode projects and workspaces, Swift packages,
CocoaPods, XcodeGen. Not yet: signing, TestFlight, React Native, Flutter.
Full guide: https://api.macbox.build/guide/ · Questions: [Discord](https://discord.gg/zukPCGsTnw)
## This repository
The open parts of macbox: the Claude Code plugin (`.claude-plugin/`, `skills/`, `.mcp.json`),
the `macbox` npm package (`npm/`, which downloads the CLI from api.macbox.build and checks
it against the published SHA256SUMS), and the MCP Registry entry (`server.json`). The macbox
service itself runs on our Macs.
MIT licensed.
What people ask about macbox-agent
What is opslane/macbox-agent?
+
opslane/macbox-agent is mcp servers for the Claude AI ecosystem. A real Mac for your coding agent: build, test, run, and tap through iOS apps from any machine It has 0 GitHub stars and its last recorded update is dated 2026-10-06.
How do I install macbox-agent?
+
You can install macbox-agent by cloning the repository (https://github.com/opslane/macbox-agent) or following the README instructions on GitHub. ClaudeWave also provides quick install blocks on this page.
Is opslane/macbox-agent safe to use?
+
Our security agent has analyzed opslane/macbox-agent and assigned a Trust Score of 87/100 (tier: Trusted). See the full breakdown of passed checks and flags on this page.
Who maintains opslane/macbox-agent?
+
opslane/macbox-agent is maintained by opslane. The last recorded GitHub activity is dated 2026-10-06, with 0 open issues.
Are there alternatives to macbox-agent?
+
Yes. On ClaudeWave you can browse similar mcp servers at /categories/mcp, sorted by popularity or recent activity.
Deploy macbox-agent to your cloud
Ship this repo to production in minutes. Each platform spins up its own environment with editable env vars.
Maintain this repo? Add a badge to your README
Drop the badge into your GitHub README to show it's tracked on ClaudeWave. Each badge links back to this page and reflects the live Trust Score.
[](https://claudewave.com/repo/opslane-macbox-agent)<a href="https://claudewave.com/repo/opslane-macbox-agent"><img src="https://claudewave.com/api/badge/opslane-macbox-agent" alt="Featured on ClaudeWave: opslane/macbox-agent" width="320" height="64" /></a>More MCP Servers
Fair-code workflow automation platform with native AI capabilities. Combine visual building with custom code, self-host or cloud, 400+ integrations.
User-friendly AI Interface (Supports Ollama, OpenAI API, ...)
An open-source AI agent that brings the power of Gemini directly into your terminal.
Real-time global intelligence dashboard. AI-powered news aggregation, geopolitical monitoring, and infrastructure tracking in a unified situational awareness interface
🕷️ An adaptive Web Scraping framework that handles everything from a single request to a full-scale crawl! Don't be shy, join here: https://discord.gg/EMgGbDceNQ and follow here for daily tips and tricks: https://x.com/Scrapling_dev
The fastest path to AI-powered full stack observability, even for lean teams.