Agno is a lightweight library for building multi-modal agents, designed with simplicity, speed, and flexibility in mind. It avoids complex graphs and chains, offering a pure Python experience for effortless development. Optimized for performance, Agno enables blazing-fast agents with minimal memory usage.

With Langtrace, you can gain deep visibility into your agents’ operations, including tool calls, memory operations, reasoning steps, and model interactions.

​
Setup

  1. Install the Langtrace’s SDK and initialize the SDK in your code.
pip install langtrace-python-sdk
  1. Install the Agno Library.
pip install agno
  1. Setup environment variables:
export LANGTRACE_API_KEY=YOUR_LANGTRACE_API_KEY
export OPENAI_API_KEY=YOUR_OPENAI_API_KEY # this is assuming you're using OPENAI for inference

​
Usage

Initialize Langtrace before creating your Phidata agent:

Python
from langtrace_python_sdk import langtrace  # Must precede other imports
from langtrace_python_sdk.utils.with_root_span import with_langtrace_root_span

from agno.agent import Agent
from agno.models.openai import OpenAIChat
from agno.tools.yfinance import YFinanceTools

langtrace.init()

Create and run the agent:


def analyze_stock(symbol: str):
    finance_agent = Agent(
        name="Finance Agent",
        model=OpenAIChat(id="gpt-4o"),
        tools=[
            YFinanceTools(
                stock_price=True,
                analyst_recommendations=True,
                company_info=True,
                company_news=True
            )
        ],
        instructions=["Use tables to display data"],
        show_tool_calls=True,
        markdown=True,
    )

    finance_agent.print_response(
        f"Analyze {symbol}. Show me:\n" +
        "1. Current stock price\n" +
        "2. Analyst recommendations\n" +
        "3. Latest company news\n" +
        "Present the information in a clear, organized format.",
        stream=True
    )

# Run the analysis
analyze_stock("NVDA")

​
What’s being traced?

With Langtrace, the following operations are automatically traced:

  1. Agent Run Operations:

    • Input/output content
    • Run duration and timestamps
    • Stream events for streaming responses

  2. Tool Calls:

    • Function name and parameters
    • Execution time
    • Return values and errors

  3. Memory Operations:

    • Memory updates
    • Chat history retrievals
    • User memory creation

View all these trace details in the Langtrace dashboard:

​
Resources