* refactor(settings): move spell check languages to constants and add type
Add Slovak language option and define SpellCheckOption type for better type safety
* fix(settings): disable spell check selector on Mac platforms
The spell check selector should not be shown on Mac platforms as it's not supported. This change adds a platform check to hide the selector when running on macOS.
* fix(prompts): clarify language detection rules for edge cases
Update LANG_DETECT_PROMPT to explicitly handle cases where the input text describes a language but is written in a different language. Add examples to illustrate the expected behavior.
* fix(prompts): correct language code mapping for Chinese input
Update the language detection prompt to properly map '英语' to 'zh-cn' instead of 'en-us' since it's a Chinese word
* Initial plan
* Fix custom parameters placement for Vercel AI Gateway
For AI Gateway provider, custom parameters are now placed at the body level
instead of being nested inside providerOptions.gateway. This fixes the issue
where parameters like 'tools' were being incorrectly added to
providerOptions.gateway when they should be at the same level as providerOptions.
Fixes#4197
Co-authored-by: DeJeune <67425183+DeJeune@users.noreply.github.com>
* Revert "Fix custom parameters placement for Vercel AI Gateway"
This reverts commit b14e48dd78.
* fix: rename 'ai-gateway' to 'gateway' across the codebase and update related configurations
* fix: resolve PR review issues for custom parameters field
- Fix Migration 174: use string literal 'ai-gateway' instead of non-existent constant for historical compatibility
- Fix Migration 180: update model.provider references to prevent orphaned models when renaming provider ID
- Add logging in mapVertexAIGatewayModelToProviderId when unknown model type is encountered
- Replace `any` with `Record<string, unknown>` in buildAIGatewayOptions return type for better type safety
- Add gateway mapping to getAiSdkProviderId mock in options.test.ts to match production behavior
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
* chore: version
* fix(options): enhance custom parameters handling for proxy providers
* fix(options): add support for cherryin provider with custom parameters handling
* chore
---------
Co-authored-by: copilot-swe-agent[bot] <198982749+Copilot@users.noreply.github.com>
Co-authored-by: DeJeune <67425183+DeJeune@users.noreply.github.com>
Co-authored-by: suyao <sy20010504@gmail.com>
Co-authored-by: Claude <noreply@anthropic.com>
* fix: update deepseek dependency to version 1.0.31 and improve provider creation logging
* chore
* feat: deepseek official hybrid infer
* fix: deepseek-v3.2-speciale tooluse and reasoning
* fix: 添加固定推理模型支持并更新相关逻辑
* refactor: simplify logic
* feat: aihubmix
* all system_providers
* feat: cherryin
* temp fix
* fix: address PR review feedback for DeepSeek v3.2 implementation
- Add default case in buildCherryInProviderOptions to fallback to genericProviderOptions
- Add clarifying comment for switch fall-through in reasoning.ts
- Add comprehensive test coverage for isFixedReasoningModel (negative cases)
- Add test coverage for new provider whitelist (deepseek, cherryin, new-api, aihubmix, sophnet, dmxapi)
- Add test coverage for isDeepSeekHybridInferenceModel prefix patterns
- Verify function calling logic works correctly via regex matching after removing provider-based checks
- Use includes() for deepseek-chat matching to support potential variants
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
* fix: remove unnecessary fall-through case for unknown providers in getReasoningEffort
---------
Co-authored-by: Claude <noreply@anthropic.com>
* fix(a11y): improve screen reader support with aria-label attributes
Add aria-label attributes to all interactive buttons and toolbar elements
to improve accessibility for screen reader users (NVDA, etc.).
Changes:
- Add aria-label with i18n translations to all ActionIconButton components
- Add role="button", tabIndex, and keyboard handlers for non-semantic elements
- Fix hardcoded English aria-labels in WindowControls to use i18n
- Add aria-pressed for toggle buttons to indicate state
- Add aria-expanded for expandable menus
- Add aria-disabled for disabled buttons
Components updated:
- SendMessageButton, CopyButton, SelectionToolbar
- CodeToolbar, RichEditor toolbar, MinimalToolbar
- WindowControls
- 12 Inputbar tool buttons (WebSearch, Attachment, KnowledgeBase, etc.)
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
* fix(a11y): enhance accessibility in CodeToolbar snapshot
Added aria-label, role, and tabindex attributes to improve screen reader support for interactive elements in the CodeToolbar component. This change aligns with ongoing efforts to enhance accessibility across the application.
---------
Co-authored-by: Claude <noreply@anthropic.com>
* fix(inputbar): block enter send while generating
- reuse unified send disable state for keyboard and button
- prevent enter sending when loading or searching
* refactor: optimize InputbarCore component's useHotkeys hook
Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>
* refactor(InputbarCore): rename cannotSend to noContent for clarity
---------
Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>
* fix(apiServer): use 127.0.0.1 instead of localhost for better compatibility
- Change default host from localhost to 127.0.0.1 in config and settings
- Add buildApiServerUrl helper to properly construct API server URLs
- Update OpenAPI documentation server URL
- Update test files to use 127.0.0.1
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
* fix(migration): migrate existing localhost config to 127.0.0.1
- Add migration 180 to automatically update localhost to 127.0.0.1
- Handle both plain host and hosts with http/https protocol
- Increment store version to 180
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
* refactor(apiServer): simplify buildApiServerUrl implementation
- Remove complex URL parsing and protocol handling
- Use simple string concatenation for URL building
- Assume http protocol since API server is local
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
* refactor: remove buildApiServerUrl helper and simplify migration
- Remove buildApiServerUrl helper function
- Use 127.0.0.1 directly in URL construction
- Simplify migration 180 to unconditionally set host to 127.0.0.1
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
* refactor(apiServer): fix critical bugs and improve code structure
🔴 Critical Fixes:
- Fix config.ts to use stored host value instead of ignoring it
- Fix hardcoded 127.0.0.1 URLs to use apiServerConfig.host
🟡 Improvements:
- Extract API_SERVER_DEFAULTS to shared constants in packages/shared/config/constant.ts
- Apply consistent fallback pattern using API_SERVER_DEFAULTS.HOST and API_SERVER_DEFAULTS.PORT
- Update all imports to use shared constants across main and renderer processes
Files changed:
- packages/shared/config/constant.ts: Add API_SERVER_DEFAULTS constants
- src/main/apiServer/config.ts: Use stored host with fallback
- src/main/apiServer/middleware/openapi.ts: Use constants
- src/renderer/src/pages/settings/ToolSettings/ApiServerSettings/ApiServerSettings.tsx: Use config host and constants
- src/renderer/src/store/settings.ts: Use constants in initial state
- src/renderer/src/store/migrate.ts: Use constants in migration
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
* update
* fix(apiServer): use relative URL in OpenAPI spec for better compatibility
- Change server URL from hardcoded defaults to relative path '/'
- This ensures Swagger UI "Try it out" works correctly regardless of configured host/port
- Remove unused API_SERVER_DEFAULTS import
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
---------
Co-authored-by: Claude <noreply@anthropic.com>
refactor(feishu-notify): simplify issue card layout by removing redundant elements
Remove unnecessary div elements and consolidate title display into the card header
* fix: update Inputbar components to support dynamic textarea height adjustment
* fix: align drag handler maxHeight with hook configuration (500px)
- Update hardcoded maxHeight from 400 to 500 in InputbarCore drag handler
- This ensures consistency with useTextareaResize hook maxHeight parameter
- Resolves PR comment about maxHeight inconsistency between hook and drag handler
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
---------
Co-authored-by: Claude <noreply@anthropic.com>
* fix(settings): fix wrong type caused by as assertion and migration
Add migration step 180 to properly handle 'undefined' string values in OpenAI settings
Update selector components to use value conversion helpers for summaryText and verbosity
* feat(models): add null as supported verbosity level for OpenAI models
Update model utils and types to include null as a valid verbosity level option alongside undefined. This provides more flexibility in controlling verbosity behavior, with null representing an explicit "off" state. Tests and UI components are updated to reflect this change.
* fix(verbosity): fix wrong verbosity type definition and handling in #11281
* style: format
* fix(store): correct verbosity check in migration config
The condition was incorrectly checking for 'undefined' string instead of undefined value, and was assigning to summaryText instead of verbosity. This fixes the migration logic to properly handle the verbosity setting.
* docs(aiCore): improve comments for verbosity and summary types
Update type comments to better explain the behavior of verbosity and summary parameters in OpenAI API requests
* fix: topic name remains after deleting last topic
- Fix logic error when deleting the last topic
- Only clear messages in the last topic before the commit
- Now creates new topic before deleting the original one to ensure proper topic name update
Signed-off-by: Do1e <i@do1e.cn>
* fix(topic): integrate the same code in `handleConfirmDelete`
Signed-off-by: Do1e <i@do1e.cn>
---------
Signed-off-by: Do1e <i@do1e.cn>
* fix: improve code block copy in collapsed state with virtual scroll
- Add saveSelection mechanism to track selection across virtual scroll updates
- Implement custom copy handler to extract complete content from raw data
- Auto-expand code block when multi-line selection is detected in collapsed state
- Only enable auto-expand when codeCollapsible setting is enabled
- Add comprehensive logging for debugging selection and copy issues
Fixes issue where copying code in collapsed state would lose content from
virtualized rows that are not rendered in DOM. The solution captures
selection position (line + offset) during scroll and uses it to extract
complete content from the original source when copying.
* fix(CodeViewer): scope selection and copy to viewer container to prevent multiple blocks appearing selected\n\n- Add selectionBelongsToViewer() to ensure selection anchors are within this viewer\n- Guard saveSelection, copy handler, and selectionchange auto-expand logic\n- Avoids cross-viewer selection bleed when multiple CodeViewer instances exist on a page\n\nFollow-up to 37c2b5ecb (virtual scroll selection/copy).
* fix(CodeViewer): clear saved selection when active selection belongs to another viewer\n\n- Early-return in selectionchange handler when selection is outside this viewer\n- Complements scoping guards to avoid misleading multi-selection states
* fix(CodeViewer): change logger info to debug for selection and copy events
- Adjust logging level from info to debug for various selection and copy operations to reduce log verbosity.
- Ensure selection belongs to the current viewer before processing.
* fix(CodeViewer): remove invisible character from import statement
* fix(CodeViewer): complete useCallback deps to avoid stale closure
- saveSelection deps -> [selectionBelongsToViewer]
- handleCopy deps -> [selectionBelongsToViewer, expanded, saveSelection, rawLines]
- no behavior change; satisfy exhaustive-deps; reduce risk of stale refs
* fix(CodeViewer): improve selection handling for virtual scrolling and enhance comments
* fix(CodeViewer): handle clipboardData unavailability and remove unused ref
- Add null check for event.clipboardData to prevent silent copy failure
- When clipboardData is unavailable, fall back to browser default copy behavior
- Remove unused isRestoringSelectionRef and its dead code check
- Improve copy reliability in edge cases where clipboard API may be unavailable
* fix: update @modelcontextprotocol/sdk to v1.23.0 and enhance MCP tool schemas
* fix: add dotenv type definitions and implement parseKeyValueString utility with tests
* docs(providerConfig): improve jsdoc for formatProviderApiHost function
* refactor(aiCore): improve provider handling with adaptProvider function
Introduce adaptProvider to centralize provider transformations and replace direct usage of handleSpecialProviders and formatProviderApiHost. This improves maintainability and provides consistent behavior across all provider usage scenarios.
* refactor(ProviderSettings): simplify api host formatting logic by using adaptProvider
Replace multiple format functions with a single adaptProvider utility to centralize host formatting logic and improve maintainability
* feat(api): add withoutTrailingSharp utility and update formatApiHost
add utility function to remove trailing # from URLs and update formatApiHost to use it
add comprehensive tests for new functionality
* feat(ProviderSetting): add help tooltip for api url selector
Add HelpTooltip component next to host selector to provide additional guidance about API URL configuration
- Deleted the old ReasoningCache class and its instance.
- Introduced CacheService for managing reasoning caches.
- Updated unified-messages service to utilize new googleReasoningCache and openRouterReasoningCache.
- Added AiSdkToAnthropicSSE adapter to handle streaming events and integrate with new cache service.
- Reorganized shared adapters to include the new AiSdkToAnthropicSSE adapter.
- Created openrouter adapter with detailed reasoning schemas for better type safety and validation.
docs: update CLAUDE.md with PR workflow details
Add critical section about Pull Request workflow requirements including reading the template, following all sections, never skipping, and proper formatting
* fix(ApiService): handle stream errors properly in checkApi
Ensure stream errors are properly caught and thrown when checking API availability
* docs(types): add type safety comment for ResponseError
Add FIXME comment highlighting weak type safety in ResponseError type
* Initial plan
* fix: set CLAUDE_CONFIG_DIR to avoid path encoding issues on Windows with non-ASCII usernames
Co-authored-by: DeJeune <67425183+DeJeune@users.noreply.github.com>
---------
Co-authored-by: copilot-swe-agent[bot] <198982749+Copilot@users.noreply.github.com>
Co-authored-by: DeJeune <67425183+DeJeune@users.noreply.github.com>
* feat(dom): extend scrollIntoView with Chromium-specific options
Add ChromiumScrollIntoViewOptions interface to support additional scroll container options
* refactor(hooks): optimize timer and scroll position hooks
- Use useMemo for scrollKey in useScrollPosition to avoid unnecessary recalculations
- Refactor useTimer to use useCallback for all functions to prevent unnecessary recreations
- Reorganize function order and improve cleanup logic in useTimer
* fix: stabilize home scroll behavior
* Update src/renderer/src/pages/home/Messages/ChatNavigation.tsx
Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>
* fix(utils/dom): add null check for element in scrollIntoView
Prevent potential runtime errors by gracefully handling falsy elements with a warning log
* fix(hooks): use ref for scroll key to avoid stale closure
* fix(useScrollPosition): add cleanup for scroll handler to prevent memory leaks
---------
Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>
* fix(i18n): remove image-generation translations and clarify endpoint type
Update English locale to specify OpenAI for image generation
Add comments to clarify image-generation endpoint type relationship
* fix(i18n): correct portuguese translations in pt-pt.json
fix(reasoning): handle reasoning effort none case properly
separate undefined and none reasoning effort cases
clean up redundant model checks and add fallback logging
Fix the issue where doubao provider could not infer models using the name field.
Refactored `isDeepSeekHybridInferenceModel` to use `withModelIdAndNameAsId`
utility function to check both model.id and model.name, avoiding duplicate
calls in `isReasoningModel`.
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-authored-by: Claude <noreply@anthropic.com>
- Updated reasoning cache to use tool-specific keys for better organization.
- Added methods to list cache keys and entries.
- Improved API version regex patterns for more accurate matching.
- Refactored API host formatting to handle leading/trailing whitespace and slashes.
- Added functions to extract and remove trailing API version segments from URLs.
