agent-device
View on GitHubMobile app automation and verification for AI coding agents. CLI, MCP server, and typed Node.js API for iOS, Android, HarmonyOS, TV, web, macOS, and Linux.
CLI, stdio MCP server, and typed Node.js API that give AI coding agents a live feedback loop: they inspect, drive, and verify iOS, Android, HarmonyOS, TV, web, macOS, and Linux apps via accessibility snapshots, then save screenshots, logs, and replay scripts for CI evidence.
Use Cases
Let a coding agent verify UI changes in a running iOS simulator or Android emulatorCapture screenshots, video, logs, and traces as reviewable PR evidenceReproduce a crash and collect the logs leading up to itRecord a manual flow as a replayable .ad script and run it in CICheck React Native re-render and performance regressionsDrive app automation from Claude Code, Codex, Cursor, Windsurf, or Cline over MCPBuild custom QA agents on the typed Node.js client or AI SDKCoordinate device ownership across parallel agent git worktreesRun mobile tests on remote device clouds (BrowserStack, AWS Device Farm, Limrun)Export runs as strict Maestro YAML for existing test suites
Built With
- Language
- TypeScript
- Frameworks
- MCP · Vercel AI SDK · React Native · Expo · Flutter · Node.js · XCUITest · ADB/HDC · Maestro · Vitest · pnpm · Appium
Tags
mcp · mcp-server · mobile-automation · ios · android · harmonyos · react-native · flutter · e2e-testing · ai-agents · coding-agents · accessibility-snapshots · screenshot-evidence · ci-cd · adb · device-cloud