F1xkex0z/investment_Agent_langgraph_crewai
🤖 Collaborate using AI agents for comprehensive A-share investment analysis, providing timely insights and risk management support.
What's novel
🤖 Collaborate using AI agents for comprehensive A-share investment analysis, providing timely insights and risk management support.
Code Analysis
12 files read · 3 roundsA multi-agent LangGraph workflow that analyzes A-share (Chinese stock) investments by combining rule-based technical/fundamental/valuation/risk analysis with LLM-powered debate and final portfolio decisions, served via a FastAPI backend.
Strengths
Genuine implementation depth in technical analysis (MACD, RSI, Bollinger, ADX, Ichimoku, 5 weighted strategies), real DCF and Owner Earnings valuation formulas, and a well-structured multi-agent pipeline with meaningful risk management (VaR, stress testing, position sizing). The LangGraph state machine with parallel branches and a bull/bear debate pattern is a thoughtful architecture for investment analysis.
Weaknesses
The README is misleadingly generic (describes a downloadable desktop app with .exe/.dmg installers that don't exist), there are zero tests, the bull/bear researchers are simplistic rule-based templates rather than true LLM-driven analysis, and the codebase has significant duplication between the src/ and crewai_system/ directories with inconsistent patterns. Heavy reliance on akshare (Chinese data source) limits portability, and the LLM integration is fragile with manual JSON parsing and bracket
Score Breakdown
Signal breakdown
Innovation
Craft
Traction
Scope
Evidence
Commits
15
Contributors
2
Files
116
Active weeks
4
Repository
Language
Python
Stars
1
Forks
0
License
NOASSERTION