How to beginner · 3 min read

How to pass context between tasks in CrewAI

Q: How to pass context between tasks in CrewAI

In CrewAI, pass context between tasks by storing relevant data in the shared context object, which is accessible across tasks. Update this context dictionary with outputs or variables from one task to be used as inputs in subsequent tasks, enabling seamless state management.

Quick answer

In CrewAI, pass context between tasks by storing relevant data in the shared context object, which is accessible across tasks. Update this context dictionary with outputs or variables from one task to be used as inputs in subsequent tasks, enabling seamless state management.

PREREQUISITES

Python 3.8+
CrewAI SDK installed (pip install crewai)
API key or environment setup for CrewAI

Setup

Install the crewai Python package and set your API key as an environment variable to authenticate your requests.

bash

pip install crewai

Step by step

This example demonstrates passing context between two tasks in a CrewAI workflow. The first task generates a greeting, and the second task uses that greeting from the shared context.

python

import os
from crewai import CrewAI, Task

# Initialize CrewAI client
client = CrewAI(api_key=os.environ["CREWAI_API_KEY"])

# Define first task: generate greeting
class GreetingTask(Task):
    def run(self, context):
        context['greeting'] = "Hello from CrewAI!"
        return context['greeting']

# Define second task: use greeting from context
class UseGreetingTask(Task):
    def run(self, context):
        greeting = context.get('greeting', 'No greeting found')
        return f"Received message: {greeting}"

# Create workflow and add tasks
workflow = client.create_workflow(name="ContextPassingWorkflow")
workflow.add_task(GreetingTask())
workflow.add_task(UseGreetingTask())

# Run workflow
result = workflow.run()
print(result)

output

Received message: Hello from CrewAI!

Common variations

You can pass complex data structures like dictionaries or lists in the context object. For asynchronous workflows, ensure tasks await context updates properly. Different CrewAI SDK versions may have slight API differences, so always check the latest docs.

python

import asyncio

class AsyncGreetingTask(Task):
    async def run(self, context):
        await asyncio.sleep(1)  # simulate async operation
        context['greeting'] = "Hello asynchronously!"
        return context['greeting']

Troubleshooting

If context data is missing in a task, verify that previous tasks correctly updated the context dictionary.
Ensure tasks run sequentially if they depend on context from prior tasks.
Check for typos in context keys to avoid key errors.

✅

Key Takeaways

Use the shared context dictionary to pass data between CrewAI tasks.
Update context in one task to make data available for subsequent tasks.
For async tasks, handle context updates with proper async/await syntax.
Always verify task execution order to maintain context integrity.
Check CrewAI SDK docs for any API changes affecting context handling.

Verified 2026-04

Verify ↗