Major improvements:
- Use period-level Sharpe ratio (range -2 to +2) instead of annualized
- Save full user prompt in decision logs for debugging
- Format complete market data (3m + 4h candles) for AI analysis
- Prevent position stacking with duplicate position checks
- Update Sharpe ratio interpretation thresholds
Market data enhancements:
- Display full technical indicators in user prompt
- Include 3-minute and 4-hour timeframe data
- Add OI (Open Interest) change and funding rate signals
Risk control:
- Block opening duplicate positions (same symbol + direction)
- Suggest close action first before opening new position
- Prevent margin usage from exceeding limits
UI improvements:
- Update multi-language translations
- Refine AI learning dashboard display
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: tinkle-community <tinklefund@gmail.com>
Problem: When TotalTrades == 0 (no completed trades), formatPerformanceFeedback
would return early, hiding Sharpe ratio and adaptive behavior recommendations
from AI's prompt. This caused AI to say "no historical data" even though Sharpe
ratio was calculated and displayed on frontend.
Solution:
- Display Sharpe ratio BEFORE checking TotalTrades
- Don't return early when TotalTrades == 0
- Always show adaptive behavior recommendations if Sharpe ratio exists
- Sharpe ratio is calculated from account equity changes, not just closed trades
Impact: AI can now properly use Sharpe ratio for self-evolution, even in early
stages before any trades are closed.
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: tinkle-community <tinklefund@gmail.com>
- Implement Sharpe ratio calculation in decision logger
- Add adaptive behavior recommendations based on Sharpe ratio
- Display Sharpe ratio in AI learning dashboard with visual indicators
- Enable AI to adjust trading strategy based on risk-adjusted returns
- Color-coded performance levels (red/yellow/green) for easy monitoring
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: tinkle-community <tinklefund@gmail.com>
- Split AI prompts into system (cacheable) and user (dynamic) messages
- Add confidence and risk_usd fields to trading decisions
- Reduce temperature to 0.5 for more stable JSON output
- Add retry mechanism for AI API calls
- Simplify candidate display (show top 10 only)
- Improve prompt structure for better AI decision quality
- Replace regex-based quote fixing with simple character replacement
- Add Twitter/X contact link to all README versions
- Remove complex regex pattern for better performance
- Fix Chinese quote mark handling in AI responses
- Add focused position strategy (max 3 concurrent positions)
- Increase position sizes for better capital efficiency
- Add historical reflection prompts for AI learning
- Fix JSON parsing with missing quotes in reasoning field
- Update position limits and leverage recommendations
- Enhance risk management with concentrated positions
- Add language context and translation system (Chinese/English)
- Enhance UI components with i18n support
- Update AILearning, EquityChart, and CompetitionPage
- Add language toggle in header
- Improve user experience with localized text
- Frontend trading records and UI enhancements
- Optimized AI prompts and decision engine
- Performance analysis and comparison features
- Binance-style UI improvements
- Remove hardcoded API URL and auth token from pool/coin_pool.go
- Update config.json.example with placeholder format
- Credentials should now be configured via config.json file
- Add English README.md with language switcher
- Add Chinese README.zh-CN.md (中文版)
- Add Ukrainian README.uk.md (Українська версія)
- Add Russian README.ru.md (Русская версия)
- Each README includes language navigation links
- Covers all core features, installation, and usage
- Multi-AI competition mode (Qwen vs DeepSeek)
- Binance Futures integration
- AI self-learning mechanism
- Professional web dashboard
- Complete risk management system
