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

# CrewAI

> Give CrewAI agents the Danube catalog as three tools

## Install

```bash theme={null}
pip install "danube[crewai]"
export DANUBE_API_KEY=...
```

## Use

```python theme={null}
from crewai import Agent, Crew, Task
from danube import DanubeClient
from danube.integrations.crewai import danube_tools

tools = danube_tools(DanubeClient())

researcher = Agent(
    role="Researcher",
    goal="Answer questions with live data",
    backstory="You find the right tool on Danube and call it.",
    tools=tools,
)

task = Task(description="What is the weather forecast for Austin this week?", expected_output="A short forecast", agent=researcher)
Crew(agents=[researcher], tasks=[task]).kickoff()
```

Three CrewAI `BaseTool`s are returned: `danube_search_tools`, `danube_describe_tool`, `danube_execute_tool`. Give every agent in a crew the same list; the search step keeps each agent's context small.

Options: `service_id`, `ready_only`, `max_results`.