---

agent-device

agent-device is a CLI for UI automation on iOS, tvOS, macOS, Android, and AndroidTV. It is designed for agent-driven workflows: inspect the UI, act on it deterministically, and keep that work session-aware and replayable.

If you know Vercel's agent-browser, this project applies the same broad idea to mobile apps and devices.

Project Goals

• Give agents a practical way to understand mobile UI state through structured snapshots.
• Keep automation flows token-efficient enough for real agent loops.
• Make common interactions reliable enough for repeated automation runs.
• Keep automation grounded in sessions, selectors, and replayable flows instead of one-off scripts.

Core Ideas

• Sessions: open a target once, interact within that session, then close it cleanly.
• Snapshots: inspect the current accessibility tree in a compact form and get current-screen refs for exploration.
• Refs vs selectors: use refs for discovery, use selectors for durable replay and assertions.
• Tests: run deterministic .ad scripts as a light e2e test suite.
• Replay scripts: save .ad flows with --save-script, replay one script with replay, or run a folder/glob as a serial suite with test. test supports metadata-aware retries up to 3 additional attempts, per-test timeouts, flaky pass reporting, and runner-managed artifacts under .agent-device/test-artifacts by default. Each attempt writes replay.ad and result.txt; failed attempts also keep copied logs and artifacts when available.
• Human docs vs agent skills: docs explain the system for people; skills provide compact operating guidance for agents.

Command Flow

The canonical loop is:

agent-device apps --platform ios
agent-device open SampleApp --platform ios
agent-device snapshot -i
agent-device press @e3
agent-device diff snapshot -i
agent-device fill @e5 "test"
agent-device press @e5
agent-device type " more" --delay-ms 80
agent-device close

In practice, most work follows the same pattern:

1. Discover the exact app id with apps if the package or bundle name is uncertain.
2. open a target app or URL.
3. snapshot -i to inspect the current screen.
4. press, fill, scroll, get, or wait using refs or selectors. On iOS and Android, default snapshot text follows the same visible-first contract: refs shown in default output are actionable now, while hidden content is surfaced as scroll/list discovery hints instead of tappable off-screen refs. Use rotate <orientation> when a flow needs a deterministic portrait or landscape state on mobile targets.
5. diff snapshot or re-snapshot after UI changes.
6. close when the session is finished.

In non-JSON mode, core mutating commands print a short success acknowledgment so agents and humans can distinguish successful actions from dropped or silent no-ops.

Where To Go Next

For people:

• Website
• Docs

For agents:

• agent-device skill
• dogfood skill
• agent-device skill on ClawHub

Install

npm install -g agent-device

agent-device now performs a lightweight background upgrade check for interactive CLI runs and, when a newer package is available, suggests a global reinstall command. Updating the package also refreshes the bundled skills/ shipped with the CLI.

Set AGENT_DEVICE_NO_UPDATE_NOTIFIER=1 to disable the notice.

On macOS, agent-device includes a local agent-device-macos-helper source package that is built on demand for desktop permission checks, alert handling, and helper-backed desktop snapshot surfaces. Release distribution should use a signed/notarized helper build; source checkouts fall back to a local Swift build.

Contributing

See CONTRIBUTING.md.

Made at Callstack

agent-device is an open source project and will always remain free to use. Callstack is a group of React and React Native geeks. Contact us at hello@callstack.com if you need any help with these technologies or just want to say hi.