- Remove all manual ReturnSchema() methods from tools
- Implement automatic schema generation using reflection
- Unify response format to flat structure with action/success/message fields
- Simplify tool implementation by removing MCPResponse embedding
- Update documentation to reflect new architecture
- Achieve ~70% code reduction while maintaining type safety
- Add ToolStartToGoal implementation with AI-driven goal automation
- Fix LLM service not initialized issue by applying global AI config to XTDriver creation
- Ensure XTDriver is created with proper AI services from the first initialization
- Add StartToGoal method to StepMobile for goal-oriented automation
- Register ToolStartToGoal in MCP server and add corresponding action type
- Add comprehensive test case for StartToGoal functionality
- Fix ReturnSchema consistency across AI tools (StartToGoal, AIAction, Finished)
- Extract AI service options in MCP argument processing
This resolves the root cause where XTDriver was created without AI services
in runStepMobileUI, ensuring only one XTDriver initialization with complete
AI service configuration.
- Move MobileAction struct from uixt package to uixt/option package
- Delete uixt/driver_action.go file as MobileAction is now in option package
- Update all import statements across the codebase to use option.MobileAction
- Update ActionTool interface to use option.MobileAction in ConvertActionToCallToolRequest method
- Maintain backward compatibility while improving package organization
- Clean up code structure by consolidating action-related types in option package
Files affected:
- server/uixt.go: Updated imports and type references
- step.go: Updated imports and ActionResult struct
- step_ui.go: Updated all MobileAction references to option.MobileAction
- uixt/mcp_server.go: Updated ActionTool interface and removed detailed comments
- uixt/mcp_server_test.go: Updated all test cases to use option.MobileAction
- uixt/mcp_tools_*.go: Updated ConvertActionToCallToolRequest method signatures
- uixt/option/action.go: Added MobileAction struct definition
- uixt/sdk.go: Updated ExecuteAction method signature
- Add detailed package-level documentation for mcp_server.go
- Create MCP_SERVER_DOCUMENTATION.md with complete implementation guide
- Create MCP_TOOLS_REFERENCE.md with quick reference for all tools
- Add extensive code comments for key structures and functions
- Document architecture, features, extension guide, and best practices
- Include usage examples and troubleshooting information
This provides complete documentation for developers to understand,
use, and extend the HttpRunner MCP server functionality.
- Replace all mapToStruct calls with parseActionOptions function
- Add parseActionOptions implementation for MCP request parameter parsing
- Remove undefined mapToStruct function that was causing compilation errors
- Standardize parameter names (fromX/fromY/toX/toY -> from_x/from_y/to_x/to_y)
- Add AntiRisk support for TapAbsXY and Drag tools
- Improve parameter validation for Drag tool
- Update corresponding test cases to match new parameter names
This fixes compilation errors and ensures all MCP tools work correctly.
- Modified ToolSwipe.ConvertActionToCallToolRequest to delegate to ToolSwipeDirection and ToolSwipeCoordinate
- Removed duplicate parameter handling logic in favor of reusing existing implementations
- Fixed linter error by removing unused variable
- Maintained backward compatibility while reducing code duplication
- All tests pass, confirming the refactoring is successful
- Added ACTION_Swipe to option/action.go for generic swipe functionality
- Implemented ToolSwipe in mcp_server.go that automatically detects parameter type:
- String params (up/down/left/right) use direction-based swipe logic
- Array params [fromX, fromY, toX, toY] use coordinate-based swipe logic
- Added comprehensive test coverage for ToolSwipe in mcp_server_test.go
- Updated tool registration to include the new generic swipe tool
- All tests pass, confirming backward compatibility with existing tools
- Removed IgnoreNotFoundError and MaxRetryTimes parameters from TapRequest, TapAbsXYRequest, and DoubleTapXYRequest structures
- Updated corresponding tool implementations to remove references to these non-existent fields
- These parameters are not applicable to coordinate-based operations as they don't involve element searching
- Only OCR/CV-based operations need these error handling parameters
This ensures that only relevant tools have the ignore_NotFoundError functionality,
making the API more consistent and avoiding confusion.
- Fixed TapByOCR and TapByCV tools to properly handle ignore_NotFoundError option
- Added option parameters to all MCP tool request structures
- Fixed ConvertActionToCallToolRequest methods to extract action options
- Added extractActionOptionsToArguments helper function for consistent option handling
- Extended fix to all MCP tools: SwipeToTapApp, SwipeToTapText, SwipeToTapTexts, TapXY, TapAbsXY
- Added comprehensive tests for option parameter handling
- Updated test expectations to match actual registered tools
This ensures that when ignore_NotFoundError is set to true, OCR/CV operations
will return nil instead of throwing errors when target elements are not found,
allowing tests to continue execution as expected.
