release: v0.5.6-beta.5 #1280
Merged
release: v0.5.6-beta.5 #1280
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
fix(ui): make update dialog scrollable
* refactor(agent): enhance system prompt with runtime context and remove browser injection * feat(renderer): add smart default maxTokens calculation with 32k cap - Add helper function to calculate safe default maxTokens - Apply 32k global limit as safety cap - Reserve space for thinking budget when reasoning is supported - Update both Chat and NewThread modes to use smart defaults - Remove hardcoded 8192 threshold logic - Add comprehensive tests for the calculation logic * fix: colada warning
* feat: add o3.fan provider * fix: update o3.fan provider apiType and official URL * chore: provider orders and defaultBaseUrl --------- Co-authored-by: zerob13 <zerob13@gmail.com>
* fix(hooks): resolve pnpm via mise * fix(hooks): resolve commit msg path
* docs: add specs for toolcall * docs: update data * feat: redesign permission and tool block * feat: add back toolcall and permission block * feat: hide permission block after granted/denied * fix: tool call response after permission * style(message): responsive layout for tool call params and response * fix(renderer): fix reasoning content flickering during streaming * fix(permission): handle denied tool calls * chore: remove spec docs
* feat: dynamic list * feat: enhance message handling and rendering - Refactor MarkdownRenderer to utilize global KaTeX and Mermaid workers. - Introduce MessageItemPlaceholder component for loading states in message lists. - Update MessageList to handle message items more efficiently and integrate placeholders. - Enhance ThinkContent with deferred rendering options for improved performance. - Implement message runtime caching for better message retrieval and management. - Modify chat store to utilize cached messages and improve message loading logic. - Add prefetching capabilities for messages to enhance user experience. - Update ChatTabView to reflect total message count and improve navigation. - Adjust playground demo to align with new message handling structure. - Extend legacy presenters with new methods for message ID retrieval and batch fetching. * feat: rewrite the message minimap with virtual scrolling * feat: srcoll to bottom first load message * feat: optimize scroll refresh --------- Co-authored-by: jiahengz <jiaheng.zhang.extern@porsche.digital>
* chore: update markdown renderer * feat: support better params and response * feat(toolcall): improve diff rendering * fix: clean code * feat: remove siliconcloud model types filter * chore: remove docs update i18n * feat: messageList access virtual scrolling (#1235) * feat: dynamic list * feat: enhance message handling and rendering - Refactor MarkdownRenderer to utilize global KaTeX and Mermaid workers. - Introduce MessageItemPlaceholder component for loading states in message lists. - Update MessageList to handle message items more efficiently and integrate placeholders. - Enhance ThinkContent with deferred rendering options for improved performance. - Implement message runtime caching for better message retrieval and management. - Modify chat store to utilize cached messages and improve message loading logic. - Add prefetching capabilities for messages to enhance user experience. - Update ChatTabView to reflect total message count and improve navigation. - Adjust playground demo to align with new message handling structure. - Extend legacy presenters with new methods for message ID retrieval and batch fetching. * feat: rewrite the message minimap with virtual scrolling * feat: srcoll to bottom first load message * feat: optimize scroll refresh --------- Co-authored-by: jiahengz <jiaheng.zhang.extern@porsche.digital> * fix: toolcall name parser * fix: review issues --------- Co-authored-by: xiaomo <wegi866@gmail.com> Co-authored-by: jiahengz <jiaheng.zhang.extern@porsche.digital>
…ion (#1266) * feat: implement skills management system with UI and backend integration - Add SkillTools class for handling skill list and control actions. - Extend ConversationsTable to include active_skills field in the database. - Create SkillsSettings.vue component for managing skills in the UI. - Update router to include skills settings route. - Add translations for skills management in English and Chinese. - Implement skills store using Pinia for state management. - Define skill-related types in shared types directory. - Integrate skill installation and uninstallation functionalities. * feat: add file selection functionality and enhance skill presenter tests - Added `selectFiles` method to `IDevicePresenter` interface for file selection with options for filters and multiple selections. - Created comprehensive tests for `SkillPresenter`, covering skill discovery, installation, activation, and deactivation. - Introduced tests for `SkillTools` to validate skill handling, including edge cases and input validation. - Mocked necessary dependencies and ensured proper integration with the event bus for skill events. * feat: document additional issues and recommendations from skills system code review * feat(skills): Refactor skill metadata prompt building and enhance skill management - Delegated skills metadata prompt construction to skillPresenter.getMetadataPrompt() to reduce code duplication. - Initialized activeSkills in default conversation settings to maintain state. - Introduced SKILL_CONFIG constants for better configuration management, including limits on file sizes and download timeouts. - Implemented checks for file sizes and content types during skill downloads to prevent memory exhaustion and ensure valid files. - Enhanced folder tree building to limit depth and skip symbolic links, preventing infinite recursion. - Updated SkillEditorSheet.vue to use YAML serialization for SKILL.md content, improving format consistency. - Simplified drag-and-drop functionality in SkillInstallDialog.vue by removing unsupported features and adding URL validation. - Added user feedback for invalid URL formats in skill installation. - Updated translations for skill name hints in both English and Chinese. * feat(skills): Add unit tests for security and tool scanning modules - Implement comprehensive unit tests for the security module, covering path safety, file size limits, permissions, and input validation. - Introduce unit tests for the ToolScanner class, validating external tool configurations, path expansion, and tool availability. - Mock filesystem interactions to isolate tests and ensure reliability. - Validate filename safety and prevent path traversal in directory scanning. - Ensure proper handling of oversized files during tool scanning. * feat(agent): gate filesystem writes * chore(i18n): add skill sync strings * chore: update i18n * feat(skills): add format adapters for Codex, Copilot User, Goose, Kilo Code, and OpenCode; update Cursor adapter to use SKILL.md format * fix(claudeCodeAdapter): specify types for id and name properties * feat(skills): Update localization files to replace "مهارتها", "Compétences", "כישורים", "技能", "Habilidades", "Навыки" with "Skills" in various languages for consistency across the application. * feat: add skills panel component and related functionality - Introduced SkillsPanel.vue to display and manage user skills. - Implemented useSkillsData composable for managing skills state and interactions. - Added useSlashMentionData composable to handle slash mention functionality for skills, prompts, and tools. - Updated MentionList.vue to support new categorization and display logic for skills and prompts. - Created SlashMention extension for TipTap editor to handle slash-triggered mentions. - Enhanced slashSuggestion.ts to manage slash mention data and filtering. - Updated i18n files for English and Chinese to include new skills-related translations. - Added new events for skill activation and deactivation. - Removed unused watchers and code related to tools and prompts from useMentionData. * feat: enhance mention handling to support slash mentions and improve content management * chore: add i18n translate * feat(skills): finish * feat: add i18n translate --------- Co-authored-by: deepinsect <deepinsect@github.com> Co-authored-by: zerob13 <zerob13@gmail.com> Co-authored-by: deepinfect <deepinfect@github.com>
Add StreamUpdateScheduler to batch and debounce streaming updates, reducing event frequency by batching updates over 200ms intervals for rendering and 600ms for database writes. Introduces stream_kind to distinguish init/delta/final events and filters events by conversation ID for performance.
* docs(spec): add tool output guardrails * fix(agent): tool output guardrails * fix(agent): standardize tool offload extension * feat: extract path to session * fix: review issue * fix: error response on renderer * feat: add read_file pagination and whitelist-based tool offload - Raise offload threshold from 3000 to 5000 characters - Add whitelist for tools that require offload (execute_command, directory_tree, etc.) - Add offset/limit params to read_file for pagination support - Auto-truncate large files at 4500 chars with metadata hint - Prevents infinite offload loop when reading offloaded files * fix: independent reasoning time for each thinking block in agent loop
Contributor
|
Important Review skippedAuto reviews are disabled on base/target branches other than the default branch. Please check the settings in the CodeRabbit UI or the You can disable this status message by setting the Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out. Comment |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
No description provided.