shoko/randebu
1
0
Fork 0
Code Issues 6 Pull Requests 1 Actions Packages Projects Releases Wiki Activity

[Backend] Chat Interface + CrewAI Integration #17

Merged
shoko merged 1 commits from fix/issue-6 into main 2026-04-08 08:37:59 +02:00
Conversation 1 Commits 1 Files Changed 4 +428 -26
shoko commented 2026-04-08 08:29:25 +02:00
Owner
Copy Link

Summary

Implement chat interface with CrewAI for natural language to strategy config parsing as described in issue #6.

Changes Made

Modified files:

  • src/backend/app/services/ai_agent/llm_connector.py - MiniMax LLM connector:

    • MiniMaxLLM class extending CrewAI LLM base class
    • MiniMaxConnector class with chat() and parse_strategy() methods
    • System prompt for strategy parsing with supported conditions

  • src/backend/app/services/ai_agent/crew.py - CrewAI crew implementation:

    • StrategyValidator class validating conditions and actions
    • StrategyExplainer class generating user-friendly explanations
    • TradingCrew class orchestrating the agents
    • Helper functions create_trading_designer_agent(), create_strategy_validator_agent(), create_strategy_explainer_agent()
    • get_trading_crew() factory function

  • src/backend/app/api/bots.py - Chat endpoint update:

    • Updated POST /api/bots/{id}/chat with CrewAI integration
    • Added conversation context management (last 10 messages)
    • Auto-updates bot strategy_config on successful parsing
    • Returns BotChatResponse with response, strategy_config, and success flag

  • src/backend/app/db/schemas.py - New schemas:

    • BotChatRequest - Request schema for chat endpoint
    • BotChatResponse - Response schema with response, strategy_config, success

Features Implemented

  1. MiniMax LLM Connector - Direct integration with MiniMax API for CrewAI
  2. Strategy Parser - Converts natural language to strategy_config JSON
  3. CrewAI Agents - Trading designer, validator, and explainer agents
  4. Strategy Validation - Validates feasibility and identifies issues
  5. Explanation Generation - User-friendly strategy explanations
  6. Conversation Context - Maintains context across chat messages

Supported Conditions (MVP)

  • price_drop - Token price drops by X%
  • price_rise - Token price rises by X%
  • volume_spike - Trading volume increases X%
  • price_level - Price crosses above/below X

Dependencies

  • Depends on issue #5 (Bot CRUD) which was already merged in PR #16

Related Issue

Fixes #6

## Summary Implement chat interface with CrewAI for natural language to strategy config parsing as described in issue #6. ### Changes Made **Modified files:** - `src/backend/app/services/ai_agent/llm_connector.py` - MiniMax LLM connector: - `MiniMaxLLM` class extending CrewAI LLM base class - `MiniMaxConnector` class with `chat()` and `parse_strategy()` methods - System prompt for strategy parsing with supported conditions - `src/backend/app/services/ai_agent/crew.py` - CrewAI crew implementation: - `StrategyValidator` class validating conditions and actions - `StrategyExplainer` class generating user-friendly explanations - `TradingCrew` class orchestrating the agents - Helper functions `create_trading_designer_agent()`, `create_strategy_validator_agent()`, `create_strategy_explainer_agent()` - `get_trading_crew()` factory function - `src/backend/app/api/bots.py` - Chat endpoint update: - Updated `POST /api/bots/{id}/chat` with CrewAI integration - Added conversation context management (last 10 messages) - Auto-updates bot strategy_config on successful parsing - Returns `BotChatResponse` with response, strategy_config, and success flag - `src/backend/app/db/schemas.py` - New schemas: - `BotChatRequest` - Request schema for chat endpoint - `BotChatResponse` - Response schema with response, strategy_config, success ### Features Implemented 1. **MiniMax LLM Connector** - Direct integration with MiniMax API for CrewAI 2. **Strategy Parser** - Converts natural language to `strategy_config` JSON 3. **CrewAI Agents** - Trading designer, validator, and explainer agents 4. **Strategy Validation** - Validates feasibility and identifies issues 5. **Explanation Generation** - User-friendly strategy explanations 6. **Conversation Context** - Maintains context across chat messages ### Supported Conditions (MVP) - `price_drop` - Token price drops by X% - `price_rise` - Token price rises by X% - `volume_spike` - Trading volume increases X% - `price_level` - Price crosses above/below X ### Dependencies - Depends on issue #5 (Bot CRUD) which was already merged in PR #16 ## Related Issue Fixes #6
shoko added 1 commit 2026-04-08 08:29:26 +02:00
feat: implement chat interface with CrewAI integration a280217254 
- Create MiniMax LLM connector for CrewAI integration
- Implement TradingCrew with trading_designer, strategy_validator, strategy_explainer
- Add strategy parsing from natural language to strategy_config JSON
- Update chat endpoint with CrewAI integration and conversation context
- Add strategy validation logic
- Add explanation generation for user-friendly responses
- Add BotChatRequest/BotChatResponse schemas

Fixes #6
han approved these changes 2026-04-08 08:37:26 +02:00
han left a comment
First-time contributor
Copy Link

lgtm

lgtm
shoko merged commit b0311bc96f into main 2026-04-08 08:37:59 +02:00
Sign in to join this conversation.
Reviewers
No Reviewers
han
Labels
Clear labels
AI/ML
AI or ML related tasks
 backend
Backend related tasks
 enhancement
Feature or enhancement
 frontend
Frontend related tasks
 infrastructure
Infrastructure or deployment
 meta
Meta issue for tracking
No Label
Milestone
No items
No Milestone
Projects
Clear projects
No project
Assignees
Clear assignees
shoko
No Assignees
2 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: shoko/randebu#17
Delete Branch "fix/issue-6"

Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?