Conversation
added 10 commits
February 16, 2026 08:52
Implement AI-powered chat tutor for language learning with: - Database: chats, chat_messages, and word_pairs tables with RLS - Backend: ChatTutorCrew with router, translation, and vocabulary agents - API: Chat endpoints for messaging, chat management, and history - Frontend: Full chat UI with sidebar, messages, and adaptive components - Features: Translation, vocabulary suggestions, domain boundary enforcement The chat agent responds to language-learning queries only and politely declines off-topic requests. Structured responses enable adaptive UI rendering for word suggestions and save confirmations.
Enhanced CLAUDE.md and AGENTS.md with: - Complete project structure and architecture overview - All API endpoints documented with examples - CrewAI agent patterns and configuration - Database schema and security notes - Frontend components, hooks, and dependencies - Debugging guide with common issues and solutions - Performance considerations and caching strategies - Environment configuration for all services - Agent-specific instructions for adding new crews This provides comprehensive reference documentation for future development and onboarding.
Implement complete flashcard management system with: - Backend API endpoints for word_pairs CRUD operations - Frontend flashcards page with list and study mode views - Interactive flashcard cards with flip animation - Study mode with progress tracking and navigation - Flashcard management (view, delete, study) - Navigation integration in sidebar Features: - My Flashcards tab: View saved vocabulary pairs with translations - Study Mode tab: Interactive flip card study sessions - Mark as Learned functionality to track progress - Progress bar showing completion percentage - Empty states with CTAs to start chatting - Flashcard deletion with confirmation The word_pairs table was created with the chat feature; this adds the UI and additional endpoints for full flashcard management.
Add documentation for using host.docker.internal when running the web container in Docker. This fixes connection issues where containers can't reach Supabase or AI service on 127.0.0.1.
- Replace non-existent IconBookOpen with IconBook - Replace non-existent IconRotateCw with IconRefresh - Replace non-existent IconGraduationCap with IconSchool - Add host.docker.internal to Vite allowedHosts for Docker browser access
- Add CrewAI tool definition for save_word_pair in tools.py - Integrate tool into translation_agent and vocabulary_agent - Pass user_id context to crew execution for tool access - Update agent backstories to mention save_word_pair capability - Enhance tool handling in run.py to inject user_id into tool calls The tool allows agents to save word pairs directly to the user's flashcard deck with duplicate detection and context sentence support.
- Add PUT /api/chats/<chat_id> endpoint to update chat title - Add renameChat function to ai-service.ts - Add renameChat to use-chat hook with optimistic update - Add inline edit UI to ChatSidebar with double-click to edit - Add keyboard shortcuts (Enter to save, Escape to cancel) Users can now rename their chats by double-clicking on the chat title or clicking the rename icon in the sidebar.
- Add RoutingDecision schema to capture router's agent selection - Create separate tasks for translation and vocabulary specialists - Implement conditional routing in run.py to execute specialist crews - Update task descriptions with proper response_type values - Add Crew and Process imports for hierarchical crew execution The router agent now analyzes user messages and delegates to: - Translation Specialist: For translation requests and grammar questions - Vocabulary Specialist: For new vocabulary word suggestions Phoenix traces show clear delegation pattern with separate spans for router decision and specialist execution.
- Clean up unused Crew, Process imports from run.py - Callback routing in crew.py creates single trace showing router → specialist chain - Agents now execute sequentially in one crew instance instead of separate instances
- Change buttons from hidden to always visible (50% opacity) - Move Rename and Delete buttons to beginning of row - Buttons become fully visible on hover
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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
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.
This pull request introduces a new
ChatTutorCrewagent configuration and exports the main crew class for easier imports. The changes are grouped below by theme.It also improves the project documentation for both the Claude and Agents guides, providing clearer instructions, detailed architecture and feature descriptions, and comprehensive onboarding for developers.
Documentation Enhancements
AGENTS.mdandCLAUDE.mdto clarify that Wordpan is a language learning application and added detailed descriptions of architecture, key features, and project structure. [1] [2]CLAUDE.md, making backend and frontend integration points explicit.AI Agent Configuration
agents.yamlconfiguration forChatTutorCrewwith three agents:router_agent,translation_agent, andvocabulary_agent, each with detailed roles, goals, and backstories focused on language learning and tool integration.ChatTutorCrewclass inai/src/crews/chat_tutor_crew/__init__.pyfor easier imports throughout the codebase.