> ## Documentation Index
> Fetch the complete documentation index at: https://www.getmaxim.ai/docs/llms.txt
> Use this file to discover all available pages before exploring further.

# Building a Financial Conversational Agent with Agno and Maxim

> Learn how to build a multi-agent financial conversational assistant using Agno for agent orchestration and Maxim for observability and tracing.

In this cookbook, you'll learn how to build a multi-agent financial conversational assistant using
[Agno](https://github.com/agno-agi/agno) for agent orchestration and
[Maxim](https://getmaxim.ai/docs) for observability and tracing. The agent can answer questions
about stocks, companies, and financial data by leveraging both web search and financial data tools,
with all interactions traced in Maxim.

## Prerequisites

* Python 3.8+
* [Agno](https://pypi.org/project/agno/) (`pip install agno`)
* [Maxim Python SDK](https://pypi.org/project/maxim-py/) (`pip install maxim-py`)
* [python-dotenv](https://pypi.org/project/python-dotenv/) (`pip install python-dotenv`)
* API keys for OpenAI and Maxim (and optionally Google Gemini)
* (Optional) Set up a `.env` file with your API keys

## 1. Import Required Libraries

```python theme={null}
from dotenv import load_dotenv
from agno.agent import Agent
# from agno.models.google.gemini import Gemini
from agno.models.openai import OpenAIChat
from agno.tools.googlesearch import GoogleSearchTools
from agno.tools.yfinance import YFinanceTools
from maxim import Maxim
from maxim.logger.agno import instrument_agno
```

## 2. Load Environment Variables and Instrument Agno with Maxim

```python {6} theme={null}
# Load environment variables from .env file
load_dotenv()

# Instrument Agno with Maxim's logger for observability
maxim = Maxim(api_key=os.getenv("MAXIM_API_KEY"))
instrument_agno(maxim.logger())
```

## 3. Define the Web Search Agent

This agent uses an LLM and Google Search tools to fetch financial information from the web.

```python theme={null}
web_search_agent = Agent(
    name="Web Agent",
    role="Search the web for information",
    # model=Gemini(id="gemini-2.0-flash-001"),
    model=OpenAIChat(id="gpt-4o"),
    tools=[GoogleSearchTools()],
    instructions="Always include sources",
    show_tool_calls=True,
    markdown=True,
)
```

## 4. Define the Finance Agent

This agent uses an LLM and YFinance tools to retrieve financial data such as stock prices, analyst
recommendations, and company information.

```python theme={null}
finance_agent = Agent(
    name="Finance Agent",
    role="Get financial data",
    # model=Gemini(id="gemini-2.0-flash-001"),
    model=OpenAIChat(id="gpt-4o"),
    tools=[YFinanceTools(stock_price=True, analyst_recommendations=True, company_info=True)],
    instructions="Use tables to display data",
    markdown=True,
)
```

## 5. Aggregate Agents into a Multi-Agent System

Combine both agents into a multi-agent system that can answer user questions by leveraging both web
search and financial data tools.

```python theme={null}
multi_ai_agent = Agent(
    team=[web_search_agent, finance_agent],
    # model=Gemini(id="gemini-2.0-flash-001"),
    model=OpenAIChat(id="gpt-4o"),
    instructions="You are a helpful financial assistant. Answer user questions about stocks, companies, and financial data.",
    show_tool_calls=True,
    markdown=True
)
```

## 6. Interactive Loop for the Financial Conversational Agent

This loop allows users to input queries and receive responses from the multi-agent system.

```python theme={null}
if __name__ == "__main__":
    print("Welcome to the Financial Conversational Agent! Type 'exit' to quit.")
    messages = []
    while True:
        print("********************************")
        user_input = input("You: ")
        if user_input.strip().lower() in ["exit", "quit"]:
            print("Goodbye!")
            break
        messages.append({"role": "user", "content": user_input})
        conversation = "\n".join([
            ("User: " + m["content"]) if m["role"] == "user" else ("Agent: " + m["content"]) for m in messages
        ])
        response = multi_ai_agent.run(
            f"Conversation so far:\n{conversation}\n\nRespond to the latest user message."
        )
        agent_reply = getattr(response, "content", response)
        print("---------------------------------")
        print("Agent:", agent_reply)
        messages.append({"role": "agent", "content": str(agent_reply)})
```

## 7. Observability with Maxim

All agent interactions, tool calls, and responses are automatically traced and can be visualized in
your [Maxim dashboard](https://app.getmaxim.ai/). This provides deep insights into agent reasoning,
tool usage, and user interactions.

<img src="https://mintcdn.com/maximai/ieTtaXix416w-43r/images/agno.gif?s=348963b90632d713933c9bcfded51908" alt="" width="1280" height="720" data-path="images/agno.gif" />

***

For more details, see the [Agno documentation](https://github.com/agno-agi/agno) and the
[Maxim Python SDK documentation](https://www.getmaxim.ai/docs/introduction/overview).

## Resources

<CardGroup cols="1">
  <Card title="Cookbook Code" icon="github" href="https://github.com/maximhq/maxim-cookbooks/blob/main/python/observability-online-eval/agno/agent.ipynb">
    Python Notebook for Agno & Maxim
  </Card>
</CardGroup>