Use when invoking android-device MCP tools to drive a real Android device or emulator over ADB (agent-fleet project) -- screen capture, tap/swipe/keyboard, app install/launch/kill, on-device shell, host device file transfer, multi-agent coordination.
Use when driving a web browser through an agent-fleet device server's agentbrowser capability (grafted Playwright MCP) -- navigating pages, reading content via accessibility snapshots, clicking/typing in web apps, end-to-end web testing, or browsing to gather information. Covers the multi-profile lease model…
Use when you must act AS THE HUMAN in a real browser through an agent-fleet device server's humanbrowser capability -- operating real logged-in accounts, changing settings/config as the person, or any task where the website must see a genuine human (zero automation traces). Covers launching the host's real Chrome and…
Use when you need DOM-level precision (exact text/aria/placeholder matching → screen coordinates) while operating the host's real Chrome with human identity and zero automation traces. humandom adds a read-only DOM locator alongside humanbrowser; all actions remain OS-level (tap/typetext/presskey). When the DOM…
Use when invoking ios-device MCP tools to drive a real iPhone or iPad over WebDriverAgent (WDA) + pymobiledevice3 (agent-fleet project) -- screen capture, tap/swipe/keyboard, app install/launch/terminate, UI hierarchy introspection, file transfer (sandbox), multi-agent coordination.
Use when invoking mac-device MCP tools to drive a Mac test machine (agent-fleet project) -- screenshot-and-click UI testing, long-running shell or zsh commands, file or process operations, AppleScript automation, or coordinating exclusive device use across multiple agents.
Use when an element you need to click is NOT in the OS accessibility tree — web pages, canvas, Electron/Flutter apps, games — so findelements/tapelement return nothing. Provides pixel-level location by visible text (OCR) or by icon image (template match), returning coordinates in the same space as tap. mac/win…
Use when invoking win-device MCP tools to drive a Windows test machine (agent-fleet project) -- screenshot-and-click UI testing, long-running shell commands, file or process operations, or coordinating exclusive device use across multiple agents.