- Bumped version of `@ai-sdk/openai-compatible` to 1.0.0-beta.8 in package.json.
- Introduced a new provider configuration for 'OpenAI Responses' in AiProviderRegistry, allowing for more flexible response handling.
- Updated type definitions to include 'openai-responses' in ProviderSettingsMap for improved type safety.
- Refactored getModelToProviderId function to return a more specific ProviderId type.
- Introduced new `WebSearchToolOutputSchema` type to standardize output from web search tools.
- Updated `webSearchTool` and `webSearchToolWithExtraction` to utilize Zod for input and output schema validation.
- Refactored tool execution logic to improve error handling and response formatting.
- Cleaned up unused type imports and comments for better code clarity.
- Revised AI SDK architecture diagram to reflect changes in component relationships, replacing PluginEngine with RuntimeExecutor.
- Updated README to highlight core features, including a refined plugin system, improved architecture design, and new built-in plugins.
- Added detailed examples for using built-in plugins and creating custom plugins, enhancing documentation for better usability.
- Included future version roadmap and related resources for user reference.
- Updated version in package.json to 1.0.0-alpha.5.
- Enhanced provider configuration validation in createProvider function for improved error handling.
- Updated the AiProviderRegistry to use direct references to creator functions for each AI provider, improving clarity and performance.
- Removed dynamic import statements for providers, simplifying the registration process and enhancing maintainability.
- Bumped version in package.json to 1.0.0-alpha.4.
- Removed deprecated dependencies from package.json and yarn.lock for improved clarity.
- Updated README to reflect changes in supported providers and installation instructions.
- Refactored provider registration and usage examples for better clarity and usability.
- Updated exports in index.ts and related files to streamline provider and model management.
- Introduced a new ModelCreator module for better encapsulation of model creation logic.
- Refactored type imports to enhance clarity and maintainability across the codebase.
- Removed deprecated provider configurations and cleaned up unused code for better performance.
- Updated version in package.json to 1.0.0-alpha.2.
- Added new path mapping for @cherrystudio/ai-core in tsconfig.web.json.
- Refactored export paths in tsdown.config.ts and index.ts for consistency.
- Cleaned up type exports in index.ts and types.ts for better organization.
Add React Native compatibility configuration to package.json, including the
react-native field and updated exports mapping. Include documentation for
React Native usage with metro.config.js setup instructions.
- Updated README to reflect the addition of a powerful plugin system and built-in web search capabilities.
- Refactored tool call handling in `ToolCallChunkHandler` to improve state management and response formatting.
- Introduced new components `MessageMcpTool`, `MessageTool`, and `MessageTools` for better handling of tool responses and user interactions.
- Updated type definitions to support new tool response structures and improved overall code organization.
- Enhanced spinner component to accept React nodes for more flexible content rendering.
- Changed the main and types entries in package.json to point to the dist directory for better output management.
- Added a new tsdown.config.ts file to define the build configuration, specifying entry points, output directory, and formats for the project.
- Included repository URL, bugs URL, and homepage in package.json for better project visibility and issue tracking.
- This update enhances the package's metadata, making it easier for users to find relevant resources and report issues.
- Updated providerSettings to allow optional 'mode' parameter for various providers, enhancing flexibility in model configuration.
- Refactored web search plugin to integrate Google search capabilities and streamline provider options handling.
- Removed deprecated code and improved type definitions for better clarity and maintainability.
- Added console logging for debugging purposes in the provider configuration process.
- Upgraded various SDK packages to their latest beta versions for improved functionality and compatibility.
- Updated `@ai-sdk/provider-utils` to version 3.0.0-beta.3.
- Adjusted dependencies in `package.json` to reflect the latest versions, including `@ai-sdk/amazon-bedrock`, `@ai-sdk/anthropic`, `@ai-sdk/azure`, and others.
- Removed outdated versions from `yarn.lock` and ensured consistency across the project.
- Added `webSearchTool` to facilitate web search functionality within the SDK.
- Updated `AiSdkToChunkAdapter` to utilize `BaseTool` for improved type handling.
- Refactored `transformParameters` to support `webSearchProviderId` for enhanced web search integration.
- Introduced new `BaseTool` type structure to unify tool definitions across the codebase.
- Adjusted imports and type definitions to align with the new tool handling logic.
* Merge memory into main
* Improvement/memory UI (#7655)
* feat: add auto-dimension detection to memory settings
- Add automatic embedding dimension detection for memory configuration
- Add toggle switch to enable/disable auto-detection (enabled by default)
- Detect dimensions by making test API call to embedding provider
- Show dimension input field only when auto-detection is disabled
- Add loading state and error handling during dimension detection
- Maintain consistency with knowledge base dimension handling
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
* ✨ feat: implement unified embedding dimensions for memory service
- Add jaison dependency for robust JSON parsing
- Normalize all embeddings to 1536 dimensions for consistency
- Improve embedding dimension logging
- Update memory processor to use jaison for better error handling
- Handle various JSON response formats in fact extraction
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
* feat: refactor MemoriesPage layout with new styled components and improved user management features
---------
Co-authored-by: Claude <noreply@anthropic.com>
* Improvement/memory UI (#7656)
Co-authored-by: Claude <noreply@anthropic.com>
* ✨ feat: add memory icon to sidebar for existing users
- Add migration version 118 to enable memory feature visibility
- Adds 'memory' icon to sidebar visible icons if not already present
- Updates store version to trigger migration for existing users
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
* fix(memory): include last user message ID in processor config
* ✨ feat(memory): enhance memory settings UI and add new translations
* Enhance memory management UI: Added settings, statistics, search, actions, and user management sections to the memory page. Updated translations for multiple languages to include new UI elements. Refactored component structure for improved layout and readability.
* feat: add i18n
* ui: Enhance memory modals and UI
* refactor(memory): replace direct message calls with window.message for error and success notifications
* fix: eslint error
* feat(memory): enhance memory restoration logic and queries
- Updated MemoryService to restore deleted memories instead of inserting new ones if a memory with the same hash exists.
- Added new SQL queries to check for deleted memories and restore them.
- Improved logging for memory restoration and embedding generation.
- Refactored related API service methods to handle updated memory processing logic.
* refactor: update memory configuration to use ApiClient structure
- Refactored memory-related services and components to utilize the new ApiClient structure for embedding and reranking models.
- Updated constructors and method signatures across multiple files to accept embedApiClient and rerankApiClient parameters.
- Enhanced memory settings UI to reflect changes in memory configuration management.
- Improved type definitions for KnowledgeBaseParams and MemoryConfig to align with the new structure.
* ui: improve user interface for adding new users in memory page
- Enhanced the button for adding new users by incorporating an icon and adjusting padding for better alignment.
- Updated the user selection options to ensure consistent alignment of avatars and user names.
- Refactored layout to improve overall user experience and visual consistency.
* refactor(memory): streamline MemoryProcessor usage in ApiService
- Removed the singleton instance of MemoryProcessor and instantiated it directly within the ApiService methods.
- Updated relevant methods to utilize the new instance for searching and processing memories, improving clarity and encapsulation of memory handling logic.
* chore: move knowledge dir
* fix: correct import paths in KnowledgeService.ts
* fix(Memory): memory deduplicate
* fix(Memory): memory llm provider
* fix: ci error
* fix(Memory): update fact extraction prompt to focus on personal information
* feat: Refactor memory fom sidebar to settings page
- Removed MemoryStick icon from Sidebar component.
- Updated navigation to point to the new memory settings page.
- Introduced MemoriesSettingsModal for managing memory configurations.
- Created MemorySettings component for comprehensive memory management.
- Added user management features including adding, editing, and deleting users.
- Implemented pagination and search functionality for memory items.
- Updated sidebar settings to remove memory icon and ensure proper migration.
- Adjusted Redux store settings to reflect changes in sidebar icons.
* feat: redesign memory settings page with improved UI and layout
* fix i18n
* fix: update citation titles to include memory hash and increment version number
* fix: remove unnecessary prop from KnowledgeCitation component
* feat: enhance fact extraction prompt with clearer guidelines and examples
* 🔧 feat: disable global memory by default and improve UI
- Set globalMemoryEnabled default to false for better user experience
- Remove manual localStorage handling to rely on redux-persist
- Add Beta badge to memory settings section
- Improve layout and styling of memory settings UI components
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
* Simplify external tool completion handling
* Fix whitespace in migrate config
---------
Co-authored-by: Claude <noreply@anthropic.com>
Co-authored-by: suyao <sy20010504@gmail.com>
Co-authored-by: eeee0717 <chentao020717Work@outlook.com>
Co-authored-by: kangfenmao <kangfenmao@qq.com>
Co-authored-by: 自由的世界人 <3196812536@qq.com>
- Updated `ModelConfig` to include a `mode` property for better differentiation between 'chat' and 'responses'.
- Modified `createBaseModel` to conditionally set the provider based on the new `mode` property in `providerSettings`.
- Refactored `RuntimeExecutor` to utilize the updated `ModelConfig` for improved type safety and clarity in provider settings.
- Adjusted imports in `executor.ts` and `types.ts` to align with the new model configuration structure.
- Refactored `helper.ts` to export new types `AnthropicSearchInput` and `AnthropicSearchOutput` for better integration with the web search plugin.
- Updated `index.ts` to include the new types in the exports for improved type safety.
- Modified `AiSdkToChunkAdapter.ts` to handle tool calls more flexibly by introducing a `GenericProviderTool` type, allowing for better differentiation between MCP tools and provider-executed tools.
- Adjusted `handleTooCallChunk.ts` to accommodate the new tool type structure, enhancing the handling of tool call responses.
- Updated type definitions in `index.ts` to reflect changes in tool handling logic.
- Updated `createBaseModel` to differentiate between OpenAI chat and response models.
- Introduced new utility functions for model identification: `isOpenAIReasoningModel`, `isOpenAILLMModel`, and `getModelToProviderId`.
- Improved `transformParameters` to conditionally set the system prompt based on the assistant's prompt.
- Refactored `getAiSdkProviderIdForAihubmix` to simplify provider identification logic.
- Enhanced `getAiSdkProviderId` to support provider type checks.
* feat: implement tool call progress handling and status updates
- Update MCP tool response handling to include 'pending' and 'cancelled' statuses.
- Introduce new IPC channel for progress updates.
- Enhance UI components to reflect tool call statuses, including pending and cancelled states.
- Add localization for new status messages in multiple languages.
- Refactor message handling logic to accommodate new tool response types.
* fix: adjust alignment of action tool container in MessageTools component
- Change justify-content from flex-end to flex-start to improve layout consistency.
* feat: enhance tool confirmation handling and update related components
- Introduced a new tool confirmation mechanism in userConfirmation.ts, allowing for individual tool confirmations.
- Updated GeminiAPIClient and OpenAIResponseAPIClient to include tool configuration options.
- Refactored MessageTools component to utilize new confirmation functions and improved styling.
- Enhanced mcp-tools.ts to manage tool invocation and confirmation processes more effectively, ensuring real-time status updates.
* refactor(McpToolChunkMiddleware): enhance tool execution handling and confirmation tracking
- Updated createToolHandlingTransform to manage confirmed tool calls and results more effectively.
- Refactored executeToolCalls and executeToolUseResponses to return both tool results and confirmed tool calls.
- Adjusted buildParamsWithToolResults to utilize confirmed tool calls for building new request messages.
- Improved error handling in messageThunk for tool call status updates, ensuring accurate block ID mapping.
* feat(McpToolChunkMiddleware, ToolUseExtractionMiddleware, mcp-tools, userConfirmation): enhance tool execution and confirmation handling
- Updated McpToolChunkMiddleware to execute tool calls and responses asynchronously, improving performance and response handling.
- Enhanced ToolUseExtractionMiddleware to generate unique tool IDs for better tracking.
- Modified parseToolUse function to accept a starting index for tool extraction.
- Improved user confirmation handling with abort signal support to manage tool action confirmations more effectively.
- Updated SYSTEM_PROMPT to clarify the use of multiple tools per message.
* fix(tagExtraction): update test expectations for tag extraction results
- Adjusted expected length of results from 7 to 9 to reflect changes in tag extraction logic.
- Modified content assertions for specific tag contents to ensure accurate validation of extracted tags.
* refactor(GeminiAPIClient, OpenAIResponseAPIClient): remove unused function calling configurations
- Removed the unused FunctionCallingConfigMode from GeminiAPIClient to streamline the code.
- Eliminated the parallel_tool_calls property from OpenAIResponseAPIClient, simplifying the tool call configuration.
* feat(McpToolChunkMiddleware): enhance LLM response handling and tool call confirmation
- Added notification to UI for new LLM response processing before recursive calls in createToolHandlingTransform.
- Improved tool call confirmation logic in executeToolCalls to match tool IDs more accurately, enhancing response validation.
* refactor(McpToolChunkMiddleware, ToolUseExtractionMiddleware, messageThunk): remove unnecessary console logs
- Eliminated redundant console log statements in McpToolChunkMiddleware, ToolUseExtractionMiddleware, and messageThunk to clean up the code and improve performance.
- Focused on enhancing readability and maintainability by reducing clutter in the logging output.
* refactor(McpToolChunkMiddleware): remove redundant logging statements
- Eliminated unnecessary logging in createToolHandlingTransform to streamline the code and enhance readability.
- Focused on reducing clutter in the logging output while maintaining error handling functionality.
* feat: enhance action button functionality with cancel and confirm options
* refactor(AbortHandlerMiddleware, McpToolChunkMiddleware, ToolUseExtractionMiddleware, messageThunk): improve error handling and code clarity
- Updated AbortHandlerMiddleware to skip abort status checks if an error chunk is received, enhancing error handling logic.
- Replaced console.error with Logger.error in McpToolChunkMiddleware for consistent logging practices.
- Refined ToolUseExtractionMiddleware to improve tool use extraction logic and ensure proper handling of tool_use tags.
- Enhanced messageThunk to include initialPlaceholderBlockId in block ID checks, improving error state management.
* refactor(ToolUseExtractionMiddleware): enhance tool use parsing logic with counter
- Introduced a toolCounter to track the number of tool use responses processed.
- Updated parseToolUse function calls to include the toolCounter, improving the extraction logic and ensuring accurate response handling.
* feat(McpService, IpcChannel, MessageTools): implement tool abort functionality
- Added Mcp_AbortTool channel to handle tool abortion requests.
- Implemented abortTool method in McpService to manage active tool calls and provide logging.
- Updated MessageTools component to include an abort button for ongoing tool calls, enhancing user control.
- Modified API calls to support optional callId for better tracking of tool executions.
- Added localization strings for tool abort messages in multiple languages.
---------
Co-authored-by: Vaayne <liu.vaayne@gmail.com>
- Added `sources` array to the default web search configuration, allowing for multiple source types including 'web', 'x', and 'news'.
- This update improves the flexibility and functionality of the web search plugin.
* feat: implement local backup functionality
- Added new IPC channels for local backup operations including backup, restore, list, delete, and set directory.
- Enhanced BackupManager with methods for handling local backups and integrated auto-sync capabilities.
- Updated settings to include local backup configurations and integrated UI components for managing local backups.
- Localized new features in English, Japanese, Russian, and Chinese.
* refactor: enhance BackupManager and LocalBackupModals for improved file handling
- Updated BackupManager to specify the type of result array for better type safety.
- Refactored showBackupModal in LocalBackupModals to use useCallback and generate a more descriptive default file name based on device type and timestamp.
* refactor: update localBackupDir path in BackupManager for consistency
- Changed localBackupDir to use the temp directory instead of userData for better alignment with backup storage practices.
* refactor: enforce localBackupDir parameter in BackupManager methods
- Updated BackupManager methods to require localBackupDir as a parameter, removing fallback to a default value for improved clarity and consistency in backup operations.
- Removed the localBackupDir property from the class, streamlining the backup management process.
* fix: update localization strings for improved clarity and consistency
- Revised English, Russian, Chinese, and Traditional Chinese localization strings for better user understanding.
- Adjusted phrases related to backup and restore processes to enhance clarity.
- Standardized terminology across different languages for consistency.
* fix: update Chinese localization strings for consistency and clarity
- Revised export menu strings in zh-cn.json to improve formatting and consistency.
- Removed spaces in phrases for a more streamlined appearance.
* feat(settings): add option to disable hardware acceleration
- Introduced a new setting to allow users to disable hardware acceleration.
- Added corresponding IPC channel and configuration management methods.
- Updated UI components to reflect the new setting and prompt for app restart.
- Localized confirmation messages for hardware acceleration changes in multiple languages.
* udpate migrate
* format code
* feat(i18n): add localized error messages for backup directory selection
- Introduced new error messages for selecting a backup directory in multiple languages, including English, Japanese, Russian, Simplified Chinese, and Traditional Chinese.
- Added checks in the LocalBackupSettings component to ensure the selected directory is not the same as the application data or installation paths, and that it has write permissions.
* format
* update migrate
* refactor(LocalBackup): streamline local backup directory validation and enhance settings UI
- Removed translation dependency in LocalBackupModals for error handling.
- Added comprehensive validation for local backup directory in LocalBackupSettings, including checks for app data path, install path, and write permissions.
- Introduced a clear directory button in the settings UI to reset the local backup directory.
- Updated the auto-sync logic to account for local backup settings.
* refactor(LocalBackupManager): remove redundant error messages for invalid local backup directory
- Eliminated repeated error message calls for invalid local backup directory in the LocalBackupManager component.
- Streamlined the validation logic to enhance code clarity and maintainability.
- Cleaned up the web search plugin code by commenting out unused sections for clarity.
- Enhanced middleware handling for the OpenAI provider by wrapping the logic in a block for better readability.
- Removed redundant imports from various files to streamline the codebase.
- Added `enableWebSearch` parameter to the fetchChatCompletion function for improved functionality.
- Introduced `createXaiOptions` function for XAI provider configuration.
- Added `XaiProviderOptions` type and validation schema in `xai.ts`.
- Updated `ProviderOptionsMap` to include XAI options.
- Enhanced `webSearchPlugin` to support XAI-specific search parameters.
- Refactored helper functions to integrate new XAI options into provider configurations.
- Added `ReasoningPart`, `FilePart`, and `ImagePart` to type exports in `index.ts`.
- Refactored `transformParameters.ts` to include `enableWebSearch` option and integrate web search tools.
- Introduced new utility `getWebSearchTools` in `websearch.ts` to manage web search tool configurations based on model type.
- Commented out deprecated code in `smoothReasoningPlugin.ts` and `textPlugin.ts` for potential removal.
- Simplified the `createModel` function to directly accept the `ModelConfig` object, improving clarity.
- Updated `createBaseModel` to include `extraModelConfig` for extended configuration options.
- Introduced `executeConfigureContext` method in `PluginManager` to handle context configuration for plugins.
- Adjusted type definitions in `types.ts` to ensure consistency with the new configuration structure.
- Refactored plugin execution methods in `PluginEngine` to utilize the resolved model directly, enhancing the flow of data through the plugin system.
- Updated the `createModel` function to accept a simplified `ModelConfig` interface, enhancing clarity and usability.
- Refactored `createBaseModel` to destructure parameters for better readability and maintainability.
- Removed the `ModelCreator.ts` file as its functionality has been integrated into the factory functions.
- Adjusted type definitions in `types.ts` to reflect changes in model configuration structure, ensuring consistency across the codebase.
- Introduced a new `webSearchPlugin` to provide unified web search functionality across multiple AI providers.
- Added helper functions for adapting web search parameters for OpenAI, Gemini, and Anthropic providers.
- Updated the built-in plugin index to export the new web search plugin and its configuration type.
- Created a new `helper.ts` file to encapsulate web search adaptation logic and support checks for provider compatibility.
* feat(settings): add option to disable hardware acceleration
- Introduced a new setting to allow users to disable hardware acceleration.
- Added corresponding IPC channel and configuration management methods.
- Updated UI components to reflect the new setting and prompt for app restart.
- Localized confirmation messages for hardware acceleration changes in multiple languages.
* fix(settings): add delay before relaunching app after disabling hardware acceleration
- Introduced a 500ms delay before the application relaunches to ensure settings are applied correctly.
- This change improves user experience by allowing time for the setting to take effect before the app restarts.
* fix lint
* fix(settings): handle errors when disabling hardware acceleration
- Wrapped the hardware acceleration disabling function in a try-catch block to manage potential errors.
- Added user feedback through an error message if the operation fails, improving overall robustness.
* feat(SelectionAssistant): add macOS support and process trust handling
- Updated the selection assistant to support macOS, including new IPC channels for process trust verification.
- Enhanced the SelectionService to check for accessibility permissions on macOS before starting the service.
- Added user interface elements to guide macOS users in granting necessary permissions.
- Updated localization files to reflect macOS support and provide relevant user instructions.
- Refactored selection-related configurations to accommodate both Windows and macOS environments.
* feat(SelectionService): update toolbar window settings for macOS and Windows
- Set the toolbar window to be hidden in Mission Control and accept the first mouse click on macOS.
- Adjusted visibility settings for the toolbar window to ensure it appears correctly on all workspaces, including full-screen mode.
- Refactored the MacProcessTrustHintModal component to improve layout and styling of buttons in the modal footer.
* feat(SelectionToolbar): enhance styling and layout of selection toolbar components
* feat(SelectionService): enhance toolbar window settings and refactor position calculation
* feat(SelectionToolbar): update button padding and add last button padding for improved layout
* chore(dependencies): update selection-hook to version 1.0.2 and refine build file exclusions in electron-builder.yml
* feat(SelectionService): center action window on screen when not following toolbar
* fix(SelectionService): implement workaround to prevent other windows from bringing the app to front on macOS when action window is closed
* fix(SelectionService): refine macOS workaround to prevent other windows from bringing the app to front when action window is closed; update selection-toolbar logo padding in styles
* fix(SelectionService): implement macOS toolbar reload to clear hover status; optimize display retrieval logic
* fix(SelectionService): update macOS toolbar hover status handling by sending mouseMove event instead of reloading the window
* chore: update selection-hook dependency to version 1.0.3 in package.json and yarn.lock
* fix(SelectionService): improve toolbar visibility handling on macOS and ensure focusability of other windows when hiding the toolbar
---------
Co-authored-by: Teo <cheesen.xu@gmail.com>
