- Implement pluggable reranker architecture with HTTP API communication
- Add LocalReranker implementation for self-hosted reranking services
- Include reference Python reranker service with Docker support
- Add comprehensive UI configuration options in settings
- Implement proper error handling and fallback mechanisms
- Add extensive test coverage for all reranking functionality
- Support internationalization for all 18 languages
- Update documentation with feature overview and usage instructions
The extension now supports optional reranking of code search results
through external services, improving search relevance while maintaining
user privacy and control.
feat: increase Claude Code default max output tokens from 8k to 16k
- Changed CLAUDE_CODE_DEFAULT_MAX_OUTPUT_TOKENS from 8000 to 16000
- Users can still lower it to 8k via environment variable if needed
- Addresses issue #6125 regarding output token limits
Co-authored-by: Roo Code <roomote@roocode.com>
* feat: add prompt caching support for LiteLLM (#5791)
- Add litellmUsePromptCache configuration option to provider settings
- Implement cache control headers in LiteLLM handler when enabled
- Add UI checkbox for enabling prompt caching (only shown for supported models)
- Track cache read/write tokens in usage data
- Add comprehensive test for prompt caching functionality
- Reuse existing translation keys for consistency across languages
This allows LiteLLM users to benefit from prompt caching with supported models
like Claude 3.7, reducing costs and improving response times.
* fix: improve LiteLLM prompt caching to work for multi-turn conversations
- Convert system message to structured format with cache_control
- Handle both string and array content types for user messages
- Apply cache_control to content items, not just message level
- Update tests to match new message structure
This ensures prompt caching works correctly for all messages in a conversation,
not just the initial system prompt and first user message.
* fix: resolve TypeScript linter error for cache_control property
Use type assertion to handle cache_control property that's not in OpenAI types
* feat: Adding more settings and control over Gemini
- with topP, topK, maxOutputTokens
- allow users to enable URL context and Grounding Research
* feat: Adding parameter titles and descriptions + translation to all languages
* feat: adding more translations
* feat: adding `contextLimit` implementation from `maxContextWindow` PR + working with profile-specific thresholding
* feat: max value for context limit to model's limit + converting description and titles to settings for translation purposes
* feat: all languages translated
* feat: changing profile-specific threshold in context management setting will also change in Gemini context management
- sync between Context Management Settting <-> Gemini Context Management with regards to thresholding
* feat: max value of maxOutputTokens is model's maxTokens + adding more tests
* feat: improve unit tests and adding `data-testid` to slider and checkbox components
* fix: small changes in geminiContextManagement descriptions + minor fix
* fix: Switching from "Gemini Context Management" to "Token Management
- better naming and correct purpose
* fix: input field showed NaN -> annoying UX
* fix: Removing redundant "tokens" after the "set context limit"'s checkbox + removing the lengthy description
* fix: Changing the translation to be consistent with the english one
* fix: more translations
* fix: translations
* fix: removing contextLimit and token management related code
- due to the decision in: https://github.com/RooCodeInc/Roo-Code/issues/3717
* fix: removing `contextLimit` test and removing token management in translations
* fix: changing from `Advanced Features` to `Tools` to be consistent with Gemini docs/AI studio
* fix: adding `try-catch` block for `generateContentStream`
* feat: Include citations + improved type safety
* feat: adding citation for streams (generateContextStream)
* fix: set default values for `topP`, `topK` and `maxOutputTokens`
* fix: changing UI/UX according to the review/feedback from `daniel-lxs`
* fix: updating the `Gemini.spec.tsx` unit test
- testing when it is hidden
- testing when users click on the collapsible trigger and model configuration appears
* fix: more changes from the feedback/review from `daniel-lxs`
* fix: adding sources at the end of the stream to preserve
* fix: change the description for grounding with google search and url context
* fix: adding translations
* fix: removing redundant extra translations - a mistake made by the agent
* fix: remove duplicate translation keys in geminiSections and geminiParameters
- Fixed duplicate keys in 13 localization files (es, fr, hi, id, it, ja, ko, nl, pl, pt-BR, ru, tr, vi)
- Removed second occurrence of geminiSections and geminiParameters keys
- Kept first occurrence which contains more comprehensive descriptions
- All JSON files validated for syntax correctness
- Translation completeness verified with missing translations script
Resolves duplicate key issue identified in PR #4895
* fix: delete topK, topP and maxOutputTokens from Gemini
* fix: deleting topK, topP and maxOutputTokens from translations/locales
* fix: adjust spacing between labels and descriptions + sentence casing
* fix: adding maxOutputTokens back and removing unknown type
* fix: internalizing error Gemini error message
* fix: updating tests in Gemini and Vertex to adjust to the new error logging
* fix: address PR review feedback for Gemini tools feature
- Fix Hindi translation grammatical error in settings.json
- Internationalize 'Sources:' string and error messages in gemini.ts
- Add comprehensive error scenario tests to gemini-handler.spec.ts
- Remove unused currentModelId prop from Gemini component
- Update all locale files with new translation keys
---------
Co-authored-by: Roo Code <roomote@roocode.com>
Co-authored-by: Matt Rubens <mrubens@users.noreply.github.com>
Co-authored-by: Daniel Riccio <ricciodaniel98@gmail.com>
* feat: Add support for message queueing
* fix: address PR review feedback for message queueing feature
- Restore original ChatView tests from main branch
- Fix broken test by updating ChatTextArea mock
- Add comprehensive tests for message queueing (simplified due to mocking constraints)
- Fix race condition using useRef and setTimeout in queue processing
- Extract QueuedMessage interface to shared types.ts file
- Replace inline styles with Tailwind classes in QueuedMessages
- Add i18n support for 'Queued Messages:' text
- Add keyboard navigation for removing queued messages
- Add JSDoc for fromQueue parameter in handleSendMessage
* refactor: move QueuedMessage interface to packages/types
- Move QueuedMessage interface from local types.ts to packages/types/src/message.ts
- Update imports in ChatView.tsx and QueuedMessages.tsx to use @roo-code/types
- Remove local types.ts file to follow codebase conventions
* fix: add id field when creating queued messages
- Generate unique id using timestamp when adding messages to queue
- Fixes TypeScript error after moving QueuedMessage interface
* Stop disabling sending
* Translations
* Fix tests
* Improved styling
* Remove unused string
* Test cleanup
* fix: address message queueing issues
- Fix race condition in queue processing by re-checking queue state inside setTimeout
- Add error handling for queue operations with retry mechanism
- Replace array index with stable message.id for React keys in QueuedMessages
- Generate more unique IDs using timestamp + random component
* feat: add inline editing for queued messages
- Add ability to edit queued messages by clicking on them
- Support Enter to save and Escape to cancel edits
- Add textarea that auto-resizes based on content
- Add hover effect to indicate messages are editable
- Add translation for click to edit tooltip
* feat: add scrollbar and fix height for queued messages
- Add max-height of 300px with scrollbar to queue container
- Add flex-shrink-0 to prevent message items from being squished
- Ensure consistent height for message items when multiple messages are queued
* feat: add translations for queued message edit tooltip
- Add 'queuedMessages.clickToEdit' translation key to all 18 language files
- Provides localized tooltip text for the click-to-edit functionality
* fix: improve message queue processing reliability
- Fix race condition by removing nested setState and setTimeout
- Add retry limit (3 attempts) to prevent infinite loops
- Add proper cleanup on component unmount
- Clear queue when starting new task
- Prevent queue processing during API errors
- Fix ESLint warnings for React hooks dependencies
---------
Co-authored-by: Roo Code <roomote@roocode.com>
Co-authored-by: Daniel Riccio <ricciodaniel98@gmail.com>
Co-authored-by: Matt Rubens <mrubens@users.noreply.github.com>
* add more details
* format details better
* fix tests
* fix: address PR #6190 review feedback
- Move huggingface-models.ts to src/api/providers/fetchers/huggingface.ts
- Remove 'any' types and add proper TypeScript interfaces
- Add missing i18n keys and translations for all languages
- Replace magic numbers with named constants
- Add JSDoc documentation for HuggingFaceModel interface
- Improve error handling in API endpoint
- Update model capabilities display to match other providers
- Remove tool calling display (not used)
- Add comprehensive test coverage for new UI features
* fix: preserve HuggingFace provider details in model response
- Store raw HuggingFace models in cache to preserve provider information
- Export getCachedRawHuggingFaceModels to retrieve full model data
- Update huggingface-models.ts to return cached raw models when available
- Include provider name in model descriptions
- Always add provider-specific variants to show all available providers
- Remove console.log statements from fetcher
---------
Co-authored-by: Daniel Riccio <ricciodaniel98@gmail.com>
fix: add Hugging Face API key to SECRET_STATE_KEYS and fix validation
- Added huggingFaceApiKey to SECRET_STATE_KEYS array to ensure it's properly handled as a secret
- Fixed validateModelsAndKeysProvided to check for huggingFaceModelId when validating Hugging Face provider
- This fixes the issue where the welcome view wouldn't transition to chat view after configuring Hugging Face
* basic hugging face provider
* fetch hf models and providers
* save provider to config
* Update translations
---------
Co-authored-by: Thomas G. Lopes <26071571+TGlide@users.noreply.github.com>
InvalidClientTokenError indicates an unrecoverable state for
the session, so we need to be more exact about triggering it. A recent
Clerk outage resulted in a lot of 429 responses which should really cause
inactive-session, not a full clear to logged-out.
* feat: add configurable timeout for evals (5-10 min)
- Add timeout field to CreateRun schema with min 5, max 10, default 5
- Add timeout slider UI component to /runs/new page
- Update database schema to include timeout column in runs table
- Create migration to add timeout column with default value of 5
- Update runTask.ts to use configurable timeout from run settings
- Pass timeout parameter through the createRun action
* fix: remove unused EVALS_TIMEOUT import
* fix: add timeout field to createRun calls in copyRun test
- Added timeout: 5 to both createRun calls in copyRun.spec.ts
- This fixes the test failure caused by the new required timeout field in the runs schema
- The timeout field was added in the configurable timeout feature but the test was not updated
* fix: use configurable timeout for Redis key expiration in registerRunner
- Updated registerRunner function to accept timeoutSeconds parameter
- Modified call in runTask.ts to pass configurable timeout instead of hardcoded EVALS_TIMEOUT
- Removed unused EVALS_TIMEOUT import from redis.ts
- Ensures Redis keys remain valid for the entire duration of task execution (up to 10 minutes)
---------
Co-authored-by: Roo Code <roomote@roocode.com>
Co-authored-by: hannesrudolph <hrudolph@gmail.com>
* fix: add character limit to prevent terminal output context explosion
- Enhanced truncateOutput function to accept character limits alongside line limits
- Character limits take priority over line limits to prevent context window explosion
- Added terminalOutputCharacterLimit setting (default: 100,000 characters)
- Updated all terminal output processing to use both limits
- Added comprehensive tests for character limit functionality
Fixes#5775
* feat: add terminal output character limit setting to UI
- Add character limit slider to Terminal Settings UI (default: 50,000)
- Update ExtensionStateContext to manage character limit state
- Add validation for positive character limit values
- Add English translation and translations for all 17 supported languages
- Connect UI to backend through proper message handling
- Character limit takes precedence over line limit to prevent memory issues
* fix: update test expectations for character limit edge cases
- Fix multi-byte character test to account for JavaScript's string length behavior
- Fix newline content test to match actual slice behavior
- Tests now correctly validate the truncateOutput function's character limit handling
---------
Co-authored-by: Roo Code <roomote@roocode.com>
Co-authored-by: Daniel Riccio <ricciodaniel98@gmail.com>
* feat: add Mistral embedding provider with OpenAI Compatible Wrapper
- Implement MistralEmbedder class using OpenAI-compatible API
- Add comprehensive unit tests with 100% coverage
- Update type definitions for Mistral provider support
- Integrate Mistral option in UI components and configuration
- Add internationalization support for Mistral provider
- Fix API key storage and retrieval for embedding providers
- Update service factory to support Mistral embeddings
- Add proper error handling and validation
This implementation allows users to use Mistral's embedding models
through the existing OpenAI-compatible wrapper approach, providing
a seamless integration experience.
* feat: add Mistral embedding provider support
- Implement MistralEmbedder class with API integration
- Add Mistral models to embedding model configurations
- Update UI to include Mistral provider option
- Add comprehensive unit tests for Mistral embedder
- Update type definitions and interfaces
- Add internationalization support for Mistral provider
* fix: add missing translations for Mistral embedding provider
* fix: address PR review feedback - improve translations and add clarifying comment
* feat: add configurable delay for Go diagnostics to prevent premature error reporting
- Add diagnosticsDelayMs setting (default: 2000ms) to allow linters time to process
- Add diagnosticsEnabled setting to optionally disable diagnostic checking entirely
- Update DiffViewProvider.saveChanges() to use configurable delay before checking diagnostics
- Update all tool files (writeToFile, searchAndReplace, insertContent, applyDiff, multiApplyDiff) to pass diagnostic settings
- Add comprehensive tests for new diagnostic functionality
- Fixes issue where Go diagnostics errors were submitted to LLM before linter could clean up unused imports
Resolves#5859
* fix: add missing TypeScript type definitions for diagnostic settings
- Add diagnosticsDelayMs and diagnosticsEnabled to globalSettingsSchema
- Include properties in ExtensionState Pick type
- Add default values to EVALS_SETTINGS
- Fix VSCode mock to include DiagnosticSeverity for tests
- Resolves compilation errors in ClineProvider and webviewMessageHandler
* fix: update test mocks to support diagnostic settings in tool tests
- Add providerRef mock to insertContentTool and writeToFileTool tests
- Update mocks to include diagnosticsEnabled and diagnosticsDelayMs settings
- Fix test expectations to match new implementation with diagnostic configuration
- Resolves failing unit tests for insertContentTool.spec.ts and writeToFileTool.spec.ts
* fix: remove package-lock.json file (project uses pnpm)
* refactor: use existing writeDelayMs instead of diagnosticsDelayMs
- Remove diagnosticsDelayMs setting in favor of existing writeDelayMs
- Add min(0) validation for writeDelayMs in global settings schema
- Add error handling around delay function calls in DiffViewProvider
- Create DEFAULT_WRITE_DELAY_MS constant (1000ms) to replace repeated defaults
- Update all tool files to pass writeDelayMs instead of diagnosticsDelayMs
- Remove diagnosticsDelayMs from webview message handlers and types
- Update test files to use writeDelayMs instead of diagnosticsDelayMs
This refactoring consolidates diagnostic delay functionality to use the
existing writeDelayMs setting as requested in PR feedback.
* fix: resolve failing unit tests and TypeScript compilation errors
- Fix DiffViewProvider test to expect correct default delay (1000ms instead of 2000ms)
- Fix TypeScript type errors in ClineProvider test mock state object
- Correct terminalPowershellCounter and terminalZdotdir types to boolean
- Fix pinnedApiConfigs type from array to Record<string, boolean>
* fix: remove unrelated changes from ClineProvider.spec.ts
- Removed extensive unrelated property additions to mock state
- Kept only diagnosticsEnabled property which is related to Go diagnostics delay feature
- Removed unused DEFAULT_WRITE_DELAY_MS import
- Restored original structure and organization of mock state object
This addresses the feedback to remove unrelated changes while preserving
the necessary diagnostic functionality for the Go diagnostics delay feature.
* refactor: move DEFAULT_WRITE_DELAY_MS to packages/types/src/global-settings.ts
- Move DEFAULT_WRITE_DELAY_MS constant from src/shared/constants.ts to packages/types/src/global-settings.ts
- Update all import statements in affected files to use @roo-code/types
- Delete src/shared/constants.ts file as it is no longer needed
- Files updated:
- src/integrations/editor/DiffViewProvider.ts
- src/core/webview/ClineProvider.ts
- src/core/tools/multiApplyDiffTool.ts
- src/core/tools/applyDiffTool.ts
- src/core/tools/searchAndReplaceTool.ts
- src/core/tools/insertContentTool.ts
- src/core/tools/writeToFileTool.ts
---------
Co-authored-by: Roo Code <roomote@roocode.com>
* fix: detect Claude models by name for API protocol selection
- Modified getApiProtocol to accept modelId parameter
- Added check for 'claude' in model name (case-insensitive)
- Updated Task.ts to pass model ID to getApiProtocol
- Added comprehensive tests for the new logic
Fixes#5830
* fix: limit Claude model detection to vertex and bedrock providers only
- Modified getApiProtocol to only detect Claude models by name when provider is vertex or bedrock
- Added comprehensive unit tests for getApiProtocol function as requested in PR review
- This ensures Claude models are only auto-detected for providers that need it
