Mobilebox
Menu

Mastra on Android: phone tools for your agent.

Mastra agents can automate Android apps through Mobilebox. MCPClient connects to the hosted Mobilebox MCP server, listTools() hands the agent its phone tools, and the agent reads a cloud phone’s screen as numbered elements, then taps, types and installs apps.

TypeScript, 29 lines, the hosted MCP server and its 13 tools. Android and iPhone over one API.

Run it in three steps.

Your code owns the phone; the agent only drives it.

Read the quickstart
  1. 1

    Get a key

    Sign up with GitHub and create an API key in the dashboard. New accounts get $2 of credit.

  2. 2

    Install Mastra

    npm i @mastra/core @mastra/mcp

  3. 3

    Run the example

    node agent.ts with MOBILESANDBOX_API_KEY and OPENAI_API_KEY set. It creates a phone, gives the agent one task and deletes the phone, also when the run fails.

The whole example

model takes any Mastra model id, so the same file runs on OpenAI, Anthropic or any other provider in the model router.

typescript · agent.ts
// Mastra + Mobilebox: a Mastra agent drives a cloud Android phone over MCP.
//   npm i @mastra/core @mastra/mcp
//   MOBILESANDBOX_API_KEY=... OPENAI_API_KEY=... node agent.ts
import { Agent } from '@mastra/core/agent'
import { MCPClient } from '@mastra/mcp'

const API = 'https://mobilebox.dev'
const AUTH = { Authorization: `Bearer ${process.env.MOBILESANDBOX_API_KEY}` }

async function phones(method: string, path = '') { // your code owns the phone: create it, and always delete it
  const res = await fetch(`${API}/v1/phones${path}`, { method, headers: AUTH })
  if (!res.ok) throw new Error(`${method} /v1/phones${path}: ${res.status} ${await res.text()}`)
  return res.status === 204 ? null : res.json()
}

const phone = (await phones('POST', '?wait=ready')).id
const mcp = new MCPClient({ servers: { mobilebox: { url: new URL(`${API}/mcp`), requestInit: { headers: AUTH } } } })
try {
  const agent = new Agent({
    id: 'phone-operator',
    name: 'Phone operator',
    instructions: `You operate the Android phone ${phone} with the Mobilebox tools. ` +
      'Observe, act, and check each result on the returned screen. Don\'t create or delete phones.',
    model: process.env.MODEL ?? 'openai/gpt-5.4-mini', // any Mastra model id
    tools: await mcp.listTools(),
  })
  const result = await agent.generate('Open Settings > Display and tell me the brightness level.', { maxSteps: 20 })
  console.log(result.text)
} finally {
  await mcp.disconnect()
  await phones('DELETE', `/${phone}`)
}

Ran once against production on 5 November 2026: the agent answered “Brightness level: 83%” in 33 s, and the phone was deleted. Run log

Questions

Before you wire Mastra to a phone.

Can I use the phone tools in a Mastra workflow?

Yes. listTools() returns Mastra tools: give them to the agent a workflow step calls, and pass the phone id in the step’s input.

Does it run in serverless functions?

Yes. The MCP connection is plain HTTPS with a bearer header, so it runs wherever Mastra runs. Disconnect when the run ends, as the example does.

Can the agent install my app?

Yes. install_app takes a public APK or split-bundle URL, or your code uploads the file over REST; then the agent opens it with {"type":"open_app","package":…}. See installing APKs and split bundles.

Does it work on iPhone?

Yes. Create the phone with platform: "ios" and the agent uses the same tools. See iPhone and iPad.

What does a run like this cost?

A ready phone costs $0.01 a minute, billed per second with a one-minute minimum; boot time and parked phones are free, so a run like the one above costs about a cent. See pricing.

Read more

Using another framework? OpenAI Agents SDK, LangGraph, CrewAI, browser-use, Vercel AI SDK, Claude Agent SDK.

All integrations