Can agents use multi tool together

[emirhanyagci]

Is agents has a multi tool calling skill ,.I couldn't find any example for this situation for example i have an agent that have googlesearch tool and newspaperk4k tool the agent first should googlesearch tool then run newspaper i tried this after google search ending procces.

.

[ekzhu]

We currently don't support strict ordering of tools. You can prompt the agent with system_message for now.

Tracking #5510

[victordibia]

Also, can you share some code to illustrate your issue?

For example, consider the following basic code below with two tools, the agent calls them both and returns the answer.

from autogen_agentchat.agents import AssistantAgent
from autogen_agentchat.ui import Console
from autogen_ext.models.openai import OpenAIChatCompletionClient

# Define a model client. You can use other model client that implements
# the `ChatCompletionClient` interface.
model_client = OpenAIChatCompletionClient(
    model="gpt-4o",
    # api_key="YOUR_API_KEY",
)


# Define a simple function tool that the agent can use.
# For this example, we use a fake weather tool for demonstration purposes.
async def get_weather(city: str) -> str:
    """Get the weather for a given city."""
    return f"The weather in {city} is 73 degrees and Sunny."


# Define an AssistantAgent with the model, tool, system message, and reflection enabled.
# The system message instructs the agent via natural language.
agent = AssistantAgent(
    name="weather_agent",
    model_client=model_client,
    tools=[get_weather],
    system_message="You are a helpful assistant.",
    reflect_on_tool_use=True,
    model_client_stream=True,  # Enable streaming tokens from the model client.
)


# Run the agent and stream the messages to the console.
async def main() -> None:
    await Console(agent.run_stream(task="What is the weather in New York?"))


# NOTE: if running this inside a Python script you'll need to use asyncio.run(main()).
await main()

---------- user ----------
What is the weather in New York and the latest news?
---------- weather_agent ----------
[FunctionCall(id='call_uSeARJLZuakb7QbquvzDVVaE', arguments='{"city": "New York"}', name='get_weather'), FunctionCall(id='call_HFLNQrccUSLRT7qGfntwtm6Y', arguments='{}', name='get_news')]
---------- weather_agent ----------
[FunctionExecutionResult(content='The weather in New York is 73 degrees and Sunny.', call_id='call_uSeARJLZuakb7QbquvzDVVaE', is_error=False), FunctionExecutionResult(content='The latest news is that AI is increasingly being used in various fields.', call_id='call_HFLNQrccUSLRT7qGfntwtm6Y', is_error=False)]
---------- weather_agent ----------
The current weather in New York is 73 degrees and sunny. In the latest news, AI is increasingly being used in various fields.

[emirhanyagci]

can agents do two different tool calling (i don't mean parallel execution) like above example if we give prompt like, "first use google search tool and use newspaper4k tool with returned urls from googlesearchtool" ?

like this

(newspaper4k tool for scraping urls)

agent = AssistantAgent(
    name="search_agent",
    model_client=model_client,
    tools=[google_search_tool,newspaper4k_tool],
    system_message="Performs a Google search to find relevant URLs and Reads and extracts information from links using newspaper4k tools",
    reflect_on_tool_use=True,
    model_client_stream=True,  # Enable streaming tokens from the model client.
)

[ekzhu]

This is an example of the scenario we would like to address: #5510

[emirhanyagci]

If you need maybe i can contribute for like something predefined RoundRobinAgent