This example demonstrates how to handle different types of events during agent run streaming. It shows how to capture and process content events, tool call started events, and tool call completed events.

​
Code

cookbook/agents/other/run_response_events.py
from typing import Iterator, List

from agno.agent import (
    Agent,
    RunContentEvent,
    RunOutputEvent,
    ToolCallCompletedEvent,
    ToolCallStartedEvent,
)
from agno.models.anthropic import Claude
from agno.tools.duckduckgo import DuckDuckGoTools

agent = Agent(
    model=Claude(id="claude-sonnet-4-20250514"),
    tools=[DuckDuckGoTools()],
    markdown=True,
)
run_response: Iterator[RunOutputEvent] = agent.run(
    "Whats happening in USA and Canada?", stream=True
)

response: List[str] = []
for chunk in run_response:
    if isinstance(chunk, RunContentEvent):
        response.append(chunk.content)  # type: ignore
    elif isinstance(chunk, ToolCallStartedEvent):
        response.append(
            f"Tool call started: {chunk.tool.tool_name} with args: {chunk.tool.tool_args}"  # type: ignore
        )
    elif isinstance(chunk, ToolCallCompletedEvent):
        response.append(
            f"Tool call completed: {chunk.tool.tool_name} with result: {chunk.tool.result}"  # type: ignore
        )

print("\n".join(response))

​
Usage

1

Create a virtual environment

Open the Terminal and create a python virtual environment.
python3 -m venv .venv
source .venv/bin/activate
2

Install libraries

pip install -U agno anthropic ddgs
3

Run Agent

python cookbook/agents/other/run_response_events.py