Show HN: An AI Agent that automates workflows on your Mobile

Mobilerun launched an open-source AI agent framework that lets developers automate mobile app workflows on real Android and iOS devices in the cloud. The tool supports natural language commands, parallel execution, and integrates with LLMs like Claude and Hermes to perform tasks such as exporting Stripe payouts or managing social media accounts.

Plug Claude, Hermes, OpenClaw, or Mobilerun VA into a real mobile device to automate any app and receive SMS. Open Stripe and export this week's payouts Real phones in the cloud. Agents run apps end to end. A profile that holds up to real-world checks. Provision real Android and iOS devices on demand. Parallel execution, fleet sync, webhooks. Reads the screen. Taps, swipes, types. Drive it from API, SDK, or natural language. Real device, real network. Apps see what they expect, so your agent just keeps going. Bring your own proxy is supported on all plans. Residential proxies available as add-on. Four workloads that only run cleanly on real devices. An LLM agent with a real phone opens apps, reads screen state, runs multi-step flows. Autonomous or orchestrated. Schedule posts, publish assets, manage creator accounts. Multi-profile, multi-region, long-lived sessions. Native apps as a structured data source. Screen state and workflow output, returned as typed JSON to your stack. Drive WhatsApp Business, Telegram, and in-app comms from a real number with a real SIM. Threads that stay open across runs, across regions. The mobilerun framework powers our cloud and runs locally on any device. Open source, MIT licensed, LLM agnostic. Start local, scale to cloud with one config change. bash install $ pip install mobilerun run your first agent $ mobilerun run "Open Settings, check battery" → task completed in 3.8s · 2 actions, 0 retries scale to cloud · one config change $ mobilerun --cloud Private fleets, custom networking, eSIM provisioning, dedicated SLA. iOS and Android in your regions, behind your access controls.