omniparser-autogui-mcp
JSON →An MCP server that analyzes the screen with OmniParser to automate GUI operations.
Tools · 10
- analyze_screen Analyzes the current screen content using OmniParser
- click_element Clicks on a specific element identified by its label or coordinates
- type_text Types text into a focused input field
- press_key Presses a keyboard key or key combination
- scroll Scrolls the screen or active window in a specified direction
- get_cursor_position Returns the current cursor position on screen
- move_cursor Moves the cursor to specified coordinates
- drag_and_drop Drags from one position to another
- wait Waits for a specified duration in milliseconds
- screenshot Takes a screenshot of the current screen or specified region
Links
★ 67 GitHub stars