Datenanalyse-Agent mit Deep Agents

from deepagents import create_deep_agent
from langchain_core.tools import tool

@tool
def execute_python(code: str) -> str:
    """Execute Python code for data analysis. pandas and matplotlib are available."""
    import subprocess
    result = subprocess.run(
        ["python", "-c", code],
        capture_output=True, text=True, timeout=30,
    )
    return result.stdout or f"Error: {result.stderr}"

@tool
def read_csv_info(path: str) -> str:
    """Get info about a CSV file (columns, types, shape)."""
    import pandas as pd
    df = pd.read_csv(path)
    return f"Shape: {df.shape}\nColumns: {list(df.columns)}\nTypes:\n{df.dtypes}"

agent = create_deep_agent(
    model="anthropic:claude-sonnet-4-5-20250929",
    tools=[execute_python, read_csv_info],
    system_prompt="You are a data analyst. Explore datasets using read_csv_info, then use execute_python to run analysis code. Present findings clearly with numbers.",
)

result = agent.invoke({
    "messages": [("user", "Analyze sales_data.csv and find the top performing products by revenue")]
})
print(result["messages"][-1].content)