How to beginner · 3 min read

How to define a task in CrewAI

Q: How to define a task in CrewAI

To define a task in CrewAI, you create a task object specifying its name, description, and input/output schema using the CrewAI SDK. This task can then be registered and executed within the CrewAI platform to automate workflows.

Quick answer

To define a task in CrewAI, you create a task object specifying its name, description, and input/output schema using the CrewAI SDK. This task can then be registered and executed within the CrewAI platform to automate workflows.

PREREQUISITES

Python 3.8+
CrewAI SDK installed via pip
CrewAI API key set in environment variable CREWAI_API_KEY

Setup

Install the CrewAI SDK and set your API key as an environment variable to authenticate requests.

bash

pip install crewai-sdk

Step by step

Define a task by creating a Task object with a name, description, and input/output schema, then register it with the CrewAI client.

python

import os
from crewai_sdk import CrewAIClient, Task

# Initialize client with API key from environment
client = CrewAIClient(api_key=os.environ["CREWAI_API_KEY"])

# Define the task
my_task = Task(
    name="SummarizeText",
    description="Summarizes input text into a concise summary.",
    inputs={"text": "string"},
    outputs={"summary": "string"}
)

# Register the task
client.register_task(my_task)

# Example usage: execute the task
result = client.execute_task(
    task_name="SummarizeText",
    inputs={"text": "CrewAI helps automate AI workflows efficiently."}
)

print("Summary:", result["summary"])

output

Summary: CrewAI automates AI workflows efficiently.

Common variations

You can define tasks with different input/output types, use async execution, or switch to different CrewAI models by specifying parameters in the Task or execute_task methods.

python

import asyncio

async def run_async_task():
    result = await client.execute_task_async(
        task_name="SummarizeText",
        inputs={"text": "Async execution example in CrewAI."}
    )
    print("Async Summary:", result["summary"])

asyncio.run(run_async_task())

output

Async Summary: Async execution example in CrewAI.

Troubleshooting

If you get authentication errors, verify your CREWAI_API_KEY environment variable is set correctly.
For schema validation errors, ensure your input/output types match the expected definitions in the Task.
If task registration fails, check network connectivity and API endpoint availability.

✅

Key Takeaways

Use the CrewAI SDK to define tasks with clear input/output schemas.
Always set your API key securely via environment variables.
Leverage async task execution for improved performance in workflows.

Verified 2026-04

Verify ↗