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

# Managing Tool Calls

This example demonstrates how to use `max_tool_calls_from_history` to limit the number of tool calls included in the agent's context.

This helps manage context size and reduce token costs while still maintaining complete history in your database.

## Code

```python filter_tool_calls_from_history.py theme={null}
import random

from agno.agent import Agent
from agno.db.sqlite import SqliteDb
from agno.models.openai import OpenAIResponses


def get_weather_for_city(city: str) -> str:
    """Get weather for a city"""
    conditions = ["Sunny", "Cloudy", "Rainy", "Snowy", "Foggy", "Windy"]
    temperature = random.randint(-10, 35)
    condition = random.choice(conditions)
    return f"{city}: {temperature}°C, {condition}"


agent = Agent(
    model=OpenAIResponses(id="gpt-5.2"),
    tools=[get_weather_for_city],
    instructions="You are a weather assistant. Get the weather using the get_weather_for_city tool.",
    # Only keep 3 most recent tool calls in context
    max_tool_calls_from_history=3,
    db=SqliteDb(db_file="tmp/weather_data.db"),
    add_history_to_context=True,
    markdown=True,
)

cities = [
    "Tokyo",
    "Delhi",
    "Shanghai",
    "São Paulo",
    "Mumbai",
    "Beijing",
    "Cairo",
    "London",
]

print(
    f"{'Run':<5} | {'City':<15} | {'History':<8} | {'Current':<8} | {'In Context':<11} | {'In DB':<8}"
)
print("-" * 90)

for i, city in enumerate(cities, 1):
    run_response = agent.run(f"What's the weather in {city}?")

    # Count tool calls in context
    history_tool_calls = sum(
        len(msg.tool_calls)
        for msg in run_response.messages
        if msg.role == "assistant"
        and msg.tool_calls
        and getattr(msg, "from_history", False)
    )

    # Count tool calls from current run
    current_tool_calls = sum(
        len(msg.tool_calls)
        for msg in run_response.messages
        if msg.role == "assistant"
        and msg.tool_calls
        and not getattr(msg, "from_history", False)
    )

    total_in_context = history_tool_calls + current_tool_calls

    # Total tool calls stored in database (unfiltered)
    saved_messages = agent.get_session_messages()
    saved_tool_calls = (
        sum(
            len(msg.tool_calls)
            for msg in saved_messages
            if msg.role == "assistant" and msg.tool_calls
        )
        if saved_messages
        else 0
    )

    print(
        f"{i:<5} | {city:<15} | {history_tool_calls:<8} | {current_tool_calls:<8} | {total_in_context:<11} | {saved_tool_calls:<8}"
    )
```

## Usage

<Steps>
  <Step title="Create a Python file">
    Create `filter_tool_calls_from_history.py` with the code above.
  </Step>

  <Snippet file="create-venv-step.mdx" />

  <Step title="Install dependencies">
    ```bash theme={null}
    uv pip install -U agno openai sqlalchemy
    ```
  </Step>

  <Step title="Export your OpenAI API key">
    <CodeGroup>
      ```bash Mac/Linux theme={null}
      export OPENAI_API_KEY="your_openai_api_key_here"
      ```

      ```bash Windows theme={null}
      $Env:OPENAI_API_KEY="your_openai_api_key_here"
      ```
    </CodeGroup>
  </Step>

  <Step title="Run Agent">
    ```bash theme={null}
    python filter_tool_calls_from_history.py
    ```
  </Step>
</Steps>