# Agno

Call a search constructor and add the resulting `agno.tools.Function` to your Agno agent's `tools` list.


# Install

``` bash
pip install "serpapi-search-tools[agno]"
```

If Agno is already installed in your project, use `pip install serpapi-search-tools` instead.

``` bash
export SERPAPI_API_KEY="your-serpapi-key"
export OPENAI_API_KEY="your-openai-key"
```


# Run it

``` python
from agno.agent import Agent
from agno.models.openai import OpenAIChat
from serpapi_search_tools import shopping_search, web_search

agent = Agent(
    model=OpenAIChat(id="gpt-5.4-mini"),
    instructions="Use SerpApi when current information helps.",
    tools=[
        web_search(),
        shopping_search(),
    ],
    telemetry=False,
)
agent.print_response("Find three well-reviewed espresso grinders for beginners.")
```


# Try it

Ask: "Find three well-reviewed espresso grinders for beginners and compare the main trade-offs."


# Customize the search

``` python
products = shopping_search(
    allowed_engines=["google_shopping", "amazon"],
    default_engine="google_shopping",
    result_limit=3,
)
```

Add `products` to the agent when product comparison is the main task.