Windows automation via Clawdos API: screen capture, mouse/keyboard input, window management, file-system operations, and shell command execution. Use when th...
This skill exposes 19 tools that let you operate a Windows machine through the Clawdos REST API. It acts as a remote Windows execution interface, granting you capabilities for screen capture, mouse and keyboard input, window management, filesystem operations, and shell execution.
This skill requires a corresponding server running on your Windows host. Download and follow the setup instructions here: danzig233/clawdos
window_list and window_focus.screen_capture to understand the current UI state.get_env against your current screenshot dimensions. If they differ, provide view_width and view_height to input tools to enable automatic coordinate scaling.key_combo, type_text) or shell commands (shell_exec) over mouse operations whenever possible.mouse_move to the estimated coordinates.screen_capture to see where the cursor landed relative to the UI element.mouse_move again.mouse_click only when verified.mouse_scroll is recommended and safe for navigating long pages.fs_* operations (fs_list, fs_read, fs_write, fs_mkdir, fs_delete, fs_move) are sandboxed to the workingDirs declared in Clawdos config; path escapes return 403.shell_exec is restricted server-side; only whitelisted commands are allowed.The core Python modules implementing the OpenClaw native skill interface.
__init__.py: OpenClaw Skill entry pointclient.py: HTTP client wrapperconfig.py: Configuration modelstools.py: Tool definitionsZIP package — ready to use