AI商业研究助手

├── llm_host.py          # Main application - handles LLM interactions
├── mcp_client.py        # MCP protocol client - connects to MCP servers
├── eve_config.py        # Eve's configuration and system prompt
├── .env                 # API keys and database connection
└── requirements.txt     # Python dependencies

pip install -r requirements.txt

# Check if Node.js is installed
node --version
npx --version

# If not installed, download from: https://nodejs.org/

AIML_API_KEY=your_aiml_api_key
TAVILY_API=your_tavily_api_key
POSTGRES_URL="postgresql://user:password@localhost:5432/database"

CREATE TABLE competitors (
    id SERIAL PRIMARY KEY,
    brand_name TEXT,
    website TEXT,
    source TEXT,
    created_at TIMESTAMP DEFAULT CURRENT_TIMESTAMP
);

python llm_host.py

You: Research Nike and store their main competitors in the database

Eve: [Uses Tavily to search for Nike competitors]
     [Stores findings in PostgreSQL database]
     [Provides comprehensive analysis]

You: What competitors do we have stored?

Eve: [Queries PostgreSQL database]
     [Returns list of stored competitors]

You: Compare Adidas with our stored Nike data

Eve: [Queries database for Nike]
     [Searches web for Adidas]
     [Provides comparative analysis]