AutoGen Stock Analysis: Two Different Implementations

The Power of AutoGen's Flexibility

This repository demonstrates how AutoGen can be used to build the same stock analysis system using two different agent architectures.

Implementation 1: Stock-Specific Agents

The first implementation uses specialized agents for stock analysis, found here url:

1. Google Search Agent: Gathers recent information about companies
2. Stock Analysis Agent: Performs technical analysis on stock data
3. Report Agent: Compiles information into comprehensive reports

Implementation 2: Universal Code Generation Agents

The second implementation shows how the same functionality can be achieved using fundamental coding agents:

1. Code Generator Agent: Writes the analysis code
2. Code Executor Agent: Runs the code and handles results
3. Report Agent: Compiles findings into reports

Why This Matters

This demonstrates AutoGen's flexibility - the same task can be approached in different ways:

• Specialized agents focused on specific domains
• General-purpose agents that can handle any coding task

Setup and Usage

1. Installation:

pip install autogen-agentchat autogen-ext openai

2. Environment Setup:

# Create .env file with your API keys
OPENAI_API_KEY=your_key_here

Running the Examples

python stock_analyzer_2.py

Key Differences

Stock-Specific Agents:

• Specialized knowledge of stock analysis
• Direct market data handling
• Domain-specific reporting

Code Generation Agents:

• Generic code writing and execution
• Flexible for any programming task
• Can be repurposed for different analyses

Conclusion

A code generator, a code executor and a file manager and I will build any agent for you!!!

Magentic-one is under development and it is exactly this!!!!! websurfer, filesurfer and a code compiler

License

MIT License