- Move output truncation from TerminalManager.processOutput() to TerminalProcess
- Pass isSubagent flag and line limit to TerminalProcess constructor
- Implement truncateToFirstAndLastHalf() helper for consistent truncation
- TerminalProcess now handles truncation during output collection
- TerminalManager.processOutput() simplified to just join lines
- Remove overrideLimit and isSubagentCommand parameters from processOutput()
This refactoring centralizes output truncation logic in TerminalProcess where
output is collected, rather than post-processing in TerminalManager. This
improves separation of concerns and ensures consistent truncation behavior
across all command executions.
Remove redundant highlight import and simplify provider search results
mapping in ApiOptions component. Search results now directly map to
items without applying highlighting, reducing unnecessary dependencies
and simplifying the code.
* Add global rules and workflows to the remote config schema
Add a way for the admin to configure global cline rules and workflows for their users.
* Add GCP Vertex provider to remote config
Add config settings for the GCP Vertex provider
Better documentation in shared/net
Better docs, wire up when running standalone.
Use esbuild to statically decide which proxying we are doing.
Add the changeset.
Add the ability to mock this fetch.
* Fixes Hicap model selector dropdown display
Fixes a bug where the checkbox was showing in the model selector dropdown.
The z-index was being improperly applied to the VSCodeTextField, causing the dropdown to be obscured. This change moves the z-index styling to be applied directly to the style attribute of the VSCodeTextField component, resolving the display issue.
* Update .changeset/proud-humans-itch.md
Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>
---------
Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>
Switch Navbar actions to the shared Button component and tooptip to use the new shadcn tooltip component so the navbar render correctly and spacing matches the previous design.
Re-add the Native Tool Call checkbox setting to the FeatureSettingsSection
that was accidentally removed in PR #7126 due to merge conflicts.
The setting allows users to enable/disable native tool calling through the
API when the feature flag is enabled. The UI includes an experimental
warning and descriptive text explaining the functionality.
Changes:
- Add nativeToolCallSetting to useExtensionState destructuring
- Add conditional checkbox UI for Native Tool Call feature
- Include experimental warning and feature description
* fix(task): prevent empty assistant messages in conversation history
Add conditional check to only append assistant content to API conversation
history when the content array is not empty. This prevents unnecessary empty
assistant messages from cluttering the conversation history and potentially
causing issues with the API.
Changes:
- Wrap addToApiConversationHistory call in length check
- Add explanatory comment for the conditional logic
* add changeset
* filter mcp marketplace results based on remote config
* add info ui banner for mcp marketplace remote config state
---------
Co-authored-by: Sarah Fortune <sarah.fortune@gmail.com>
* feat: add AIhubmix provider integration
- Add AIhubmix as a new provider with full API integration
- Implement AIhubmixHandler for API interactions
- Add model fetching functionality via getAihubmixModels
- Create UI components for AIhubmix configuration
- Update proto definitions and API configuration
- Add changeset for version tracking
* docs: add AIhubmix integration documentation
* fix: add plan/act mode fields for AIhubmix and Hicap providers
- Add planModeAihubmixModelId and planModeAihubmixModelInfo
- Add actModeAihubmixModelId and actModeAihubmixModelInfo
- Add planModeHicapModelId and planModeHicapModelInfo
- Add actModeHicapModelId and actModeHicapModelInfo
- Ensures model selection works correctly in both plan and act modes
* merge
* Delete AIHUBMIX_INTEGRATION.md
* fix: bot
* Update providerUtils.ts
This fixes a flaky e2e test by adding an explicit wait for the slash
commands menu to be visible before attempting to interact with it.
Changes:
- Add data-testid to SlashCommandMenu component for test targeting
- Add waitFor() call in test to ensure menu is visible before Tab press
* fix: enable Add Header button in OpenAICompatibleProvider
Implement functionality for the Add Header button that was previously disabled. The button now:
- Enables when headers are not managed by remote configuration
- Creates new header entries with auto-incremented keys (header1, header2, etc.)
- Updates the API configuration state with the new header
This allows users to add custom headers to their OpenAI-compatible provider configuration unless restricted by organization settings.
* add changeset
* Feat:
* add hicap as provider option
* add new variable to handle hicapApiKey
* add variable to handle hicapModelId in Plan and Act mode
* get hicap available model from Hicap Endpoint
* create hicap provider (ui)
* hicap handler
* rebase main into this branch and fixing errors
* add changeset
* fix typo with hicap api key
* resolve comments from cline team on PR
* revert some delete console logs
* HicapModelPicker changed styled components for tailwind format, remove unnecessary hicapModelId migration, refreshHicapModel use setGlobalState function
* rebase main branch
* Remove redundant test code that's very confusing
* feat: add vscodeTerminalExecutionMode to TaskConfig and apply timeout
Add vscodeTerminalExecutionMode parameter to TaskConfig and thread it through ToolExecutor and Task classes. Apply default 30-second timeout to commands executed in backgroundExec mode, similar to existing yolo mode behavior.
This change enables different execution strategies for VSCode terminal commands and ensures background executions have appropriate timeout protection to prevent hanging processes.
- Add anthropic--claude-4.5-sonnet model definition to sapAiCoreModels
- Update SAP AI Core provider to handle Claude 4.5 Sonnet in anthropicModels array
- Enable caching support for Claude 4.5 Sonnet using converse-stream endpoint
- Add Claude 4.5 Sonnet to streamCompletionSonnet37 method for proper response handling
Fixes#7216
Co-authored-by: Wahaj Ahmed Khan <“wahaj.khan@sap.com”>
- Add provider parameter to captureToolUsage() and captureDiffEditFailure() methods in TelemetryService
- Update all 11 tool handlers to extract and pass provider information
- Extract provider using plan/act mode differentiation from state manager
- Update UIHelpers.ts captureTelemetry wrapper to include provider
- Update test file to include provider parameter in test calls
This enables tracking which API provider (anthropic, openai, etc.) was used for each tool execution.
* fix: react-remark rendering in ModelDescriptionMarkdown
- Add useRemark hook to properly parse and render markdown content that was removed in last git commit
- Extract props interface to ModelDescriptionMarkdownProps for better type safety
- Add useEffect to reactively update markdown when content changes
- Set fixed height (h-20) for collapsed state to improve layout consistency
- Replace raw markdown text display with processed reactContent
This change ensures markdown formatting (links, bold, italics, etc.) is correctly rendered in model descriptions instead of showing raw markdown syntax.
* changeset
* feat(hooks): Initial implementation of hooks UI using background terminal UI
* Hooks UI improvements
* Hook discovery improvements
* Separate hooks UI into separate files
* Fix failing tests
* Add missing docs
* Hooks hardening and handling edge cases
* Improvements to hooks error messaging
* Improve TaskCancel hook UI
* Simplify and improve console.log() output from hooks
* Changes from usability feedback
* test: verify linting fixes
* fix: resolve cancel functionality in PostToolUse hook using flag pattern
* fix: add cancel button support for TaskResume and UserPromptSubmit hooks
* Getting things working again
* Fixing cancel behaviors
* Trying another fix
* Demoed
* Add examples to .clinerules/hooks/ directory
* Refactored away loadTaskStateWithoutWorkflow
* Fix unsafe return in finally block
Refactored PostToolUse hook execution to avoid using return statement
in finally block, which the linter flags as unsafe. Instead of early
return, wrapped the hook logic in a conditional check for abort status.
This resolves the biome lint error:
lint/correctness/noUnsafeFinally - Unsafe usage of 'return' in finally block
* Remove unnecessary files from branch
* remove cancelHookExecution() in favor of cancelTask()
* Remove unnecessary polling pattern for hook cancellation
* Fixing cancel behavior
* Fixing cancel behavior
* Fix resume
* Sqaush commits to improve merge tool behavior
Plan out the work needed to resolve race conditions
Fix race conditions (1)
Fix race conditions (2)
Fix race conditions (3)
Fix race conditions (4)
Fix race conditions (5)
Fix unit tests
Updated implementation plan doc
Code quality improvement
Temporary logging to troubleshoot race conditions in hooks UI blocks
Fix streaming output for hooks
Remove in-progress planning file
Revert breakage
Remove unneeded arg output in PreToolUse hook UI
Fix race condition with PreToolUse and PostToolUse hooks
Fix PreToolUse attempt_completion use case
Fix ChatTextArea prompt input
Fixed part of the rersume bug
Clean-up before code review
Change cancel button to abort button
Fixed TaskCancel behavior
Fixing resume behavior (partially fixed)
Make notch arrow visible in hooks block up expand/contract toggle
Prevent cancel/abort from clearing unsent user message
Change 'Cancelled' to 'Aborted' in hook UI block
Use color styles consistent with background terminal
Fix npm compile issue
Remove unneeded hot-cold tracking from hooks
Remove unused constant as per PR feedback
Minor changes
Skip combineHookSequences() if hooks feature setting not enabled
Improve mutex pattern for general use and fix additional race condition
Refactor to reduce duplication (keep it DRY)
Refactor to reduce duplication (keep it DRY) part 2
Fix tests
Skip some tests on Windows (not yet supported)
Hooks unsupported on Windows
Fix broken unit tests
* Fix broken unit tests
* Bee and Eve fixing 'API Request...' test in e2e tests
* Add comment as per ellipsis-dev PR feedback
* Changes as per PR feedback from Evan
* Remove if that always resolves to true
* Don't use vscode API; use cross-compatible solutions instead
* Remove mention of obsolete clineAsk handler
- Add native tool calling support with configurable setting.
- Auto-approve is now always-on with a redesigned expanding menu. Settings simplified and notifications moved to General Settings.
- added zai-glm-4.6 as a Cerebras model
- Created GPT5 family specific system prompt template
- Fix: show reasoning budget slider to models with valid thinking config
- Requesty base URL, and API key fixes
- Delete all Auth Tokens when logging out
- Support for <think> tags for models that prefer that over <thinking>
Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com>
* feat: add native tool call setting with feature flag support
Add a new native tool call setting that allows users to enable/disable native tool calling functionality. The setting is controlled by both a user preference and a feature flag, providing flexibility for gradual rollout. Changes include:
- Added native_tool_call_enabled field to UpdateSettingsRequest proto
- Implemented state management for the setting in controller and storage
- Added UI toggle in feature settings section
- Updated task execution logic to check both user setting and feature flag
- Extended extension state interface to include the new setting
* Add changeset
Previously, hasThinkingBudget would return true for empty objects ({}),
causing incorrect behavior. Now checks both for truthiness and that the
object contains at least one key.
Also updates thinkingConfig assignment logic to avoid setting empty
objects and converts imports to type-only imports for better tree-shaking.
Refine TaskHeader controls so actions wrap naturally and the cost
label scales down on narrow viewports.
Update tooltip content to wrap flexibly, tweak padding,
add collision padding, and cap width on small screens.
Introduce xs and xxs breakpoint variables to support responsive styles.
* fix: set free pricing for featured models in Cline provider
Add isFree flag to featured models and ensure free models have $0 pricing when using Cline provider. This updates the OpenRouterModelPicker to automatically set input, output, and cache prices to zero for free featured models, preventing confusing info display to users.
* update
* fix
* fix: adjust mode toggle hover and colors
- add hover state to plan/act toggle buttons with toolbar hover color
- set inactive text color to text-foreground
- align toggle styling with previous design
* feat: toggle spacing fix and transparent color
* SwitchContainer to tailwind
* revert
---------
Co-authored-by: Jose R. Perez <trupix@gmail.com>
Arafatkatze