CS2 Skin Economy Backend

A powerful backend service for CS2 (Counter-Strike 2) skin economy research and analysis, built with FastAPI and LangChain.

Architecture Overview

graph TD
    A[FastAPI Server] --> B[LangChain Agent]
    B --> C[Search Tools]
    B --> D[Document Tools]
    B --> E[CS Skins Tool]
    
    C --> F[Search Utils]
    C --> G[Fallback Search]
    
    F --> H[FAISS Index]
    F --> I[Sentence Transformers]
    
    D --> J[Document Processing]
    D --> K[Pattern Analysis]
    
    E --> L[Price Analysis]
    E --> M[Market Data]

Loading

Core Components

1. FastAPI Server (main.py)

• Entry point for the application
• Handles HTTP requests and CORS
• Initializes the search engine and tools
• Manages the LangChain agent

2. Search Engine (search_utils.py & search_utils_simplified.py)

• Primary search implementation using FAISS and Sentence Transformers
• Fallback mechanism for when dependencies are missing
• Features:
  • Semantic search using embeddings
  • Fuzzy matching for similar items
  • Price-based filtering
  • Pattern recognition

3. Tools (tools.py)

• Collection of specialized tools for the LangChain agent:
  • cs_skins_tool: CS2 skin price and market data
  • search_tool: Web search capabilities
  • wiki_tool: Wikipedia information
  • save_tool: Data persistence
  • document_tool: Document analysis

4. Document Processing (document_tools.py)

• Handles document analysis and pattern recognition
• Processes structured data for the search engine
• Manages document caching and updates

Setup Instructions

1. Clone the repository:

git clone https://github.com/yourusername/2m-backend.git
cd 2m-backend

2. Create and activate a virtual environment:

python -m venv venv
source venv/bin/activate  # On Windows: venv\Scripts\activate

3. Install dependencies:

pip install -r requirements.txt

4. Set up environment variables:

cp .env.example .env
# Edit .env with your configuration

5. Start the server:

python main.py

API Endpoints

POST /query

Main endpoint for processing queries about CS2 skins and market data.

Request:

{
    "query": "What's the cheapest AK-47 skin?"
}

Response:

{
    "answer": "Detailed response about AK-47 skins...",
    "sources": ["source1", "source2"]
}

GET /

Health check endpoint.

Search Capabilities

The backend supports various types of queries:

1. Price-based queries:

   • "cheapest AK-47"
   • "AWP skins under $50"
   • "Glock-18 between $10 and $30"

2. Pattern-based queries:

   • "Doppler patterns"
   • "Fade patterns"

3. Market analysis:

   • "price trends"
   • "market statistics"

Development Guidelines

Adding New Tools

1. Create a new tool function in tools.py
2. Register it in the tools list in main.py
3. Update the agent's prompt if necessary

Modifying Search Behavior

1. Edit search_utils.py for core search functionality
2. Update search_utils_simplified.py for fallback behavior
3. Test both implementations

Error Handling

• The system includes fallback mechanisms for missing dependencies
• Errors are logged and handled gracefully
• User-friendly error messages are returned

Contributing

1. Fork the repository
2. Create a feature branch
3. Commit your changes
4. Push to the branch
5. Create a Pull Request

License

MIT License - see LICENSE file for details