> ## Documentation Index
> Fetch the complete documentation index at: https://stagehand-codex-replicate-pr-2348.mintlify.site/llms.txt
> Use this file to discover all available pages before exploring further.

# Use CrewAI to Automate Browser Tasks

> Create intelligent agents that can interact with websites and automate browser tasks using natural language instructions

This guide walks you through setting up CrewAI with Browserbase to create agents that can perform web automation tasks using natural language instructions.

## Step 1: Install Dependencies

Install the required packages for CrewAI and Stagehand integration:

```bash theme={null}
pip install stagehand-py crewai crewai-tools
```

## Step 2: Configure Environment Variables

You'll need API keys from three services:

1. **Browserbase API Key and Project ID**: Get these from your [Browserbase dashboard](https://www.browserbase.com/)
2. **LLM API Key**: Get an API key from [OpenAI](https://platform.openai.com/api-keys) or [Anthropic](https://console.anthropic.com/)

Store your API keys securely as environment variables:

```bash theme={null}
BROWSERBASE_API_KEY="your-browserbase-api-key"
BROWSERBASE_PROJECT_ID="your-browserbase-project-id"
OPENAI_API_KEY="your-openai-api-key"
ANTHROPIC_API_KEY="your-anthropic-api-key"
```

## Step 3: Create Your First Agent

Create a Python script with a basic CrewAI agent:

```python theme={null}
import os
from crewai import Agent, Task, Crew
from crewai_tools import StagehandTool
from stagehand.schemas import AvailableModel

# Get API keys from environment
browserbase_api_key = os.environ.get("BROWSERBASE_API_KEY")
browserbase_project_id = os.environ.get("BROWSERBASE_PROJECT_ID")
model_api_key = os.environ.get("OPENAI_API_KEY")  # or ANTHROPIC_API_KEY

# Initialize the StagehandTool
stagehand_tool = StagehandTool(
    api_key=browserbase_api_key,
    project_id=browserbase_project_id,
    model_api_key=model_api_key,
    model_name=AvailableModel.GPT_4O,  # or AvailableModel.CLAUDE_3_7_SONNET_LATEST
)

# Create an agent with the tool
researcher = Agent(
    role="Web Researcher",
    goal="Find and summarize information from websites",
    backstory="I'm an expert at finding information online.",
    verbose=True,
    tools=[stagehand_tool],
)
```

## Step 4: Create and Run a Task

Define a task for your agent and execute it:

```python theme={null}
# Create a task that uses the tool
research_task = Task(
    description="Go to https://www.example.com and tell me what you see on the homepage.",
    agent=researcher,
)

# Run the crew
crew = Crew(
    agents=[researcher],
    tasks=[research_task],
    verbose=True,
)

try:
    result = crew.kickoff()
    print(result)
finally:
    # Clean up resources
    stagehand_tool.close()
```

## Step 5: Run Your Script

Execute your Python script:

```bash theme={null}
python your_crew_script.py
```

## Advanced Configuration

Customize the StagehandTool behavior with additional parameters:

```python theme={null}
stagehand_tool = StagehandTool(
    api_key=browserbase_api_key,
    project_id=browserbase_project_id, 
    model_api_key=model_api_key,
    model_name=AvailableModel.CLAUDE_3_7_SONNET_LATEST,
    dom_settle_timeout_ms=5000,  # Wait longer for DOM to settle
    headless=True,  # Run browser in headless mode
    self_heal=True,  # Attempt to recover from errors
    wait_for_captcha_solves=True,  # Wait for CAPTCHA solving
    verbose=1,  # Control logging verbosity (0-3)
)
```

## Example Tasks

<Tabs>
  <Tab title="Form Submission" value="form-submission" label="Python">
    ```python theme={null}
    form_task = Task(
        description="""
        Submit a contact form:
        1. Go to https://example.com/contact
        2. Fill out the form with name 'John Doe', email 'john@example.com'
        3. Submit and confirm success
        """,
        agent=researcher,
    )
    ```
  </Tab>

  <Tab title="Data Extraction" value="data-extraction" label="Python">
    ```python theme={null}
    extraction_task = Task(
        description="""
        Extract product information:
        1. Go to the products page
        2. Extract all product names, prices, and descriptions
        3. Format as structured data
        """,
        agent=researcher,
    )
    ```
  </Tab>

  <Tab title="Multi-step Navigation" value="multi-step-navigation" label="Python">
    ```python theme={null}
    navigation_task = Task(
        description="""
        Navigate and analyze:
        1. Start at homepage
        2. Navigate to products section  
        3. Filter by 'Electronics' category
        4. Find and extract details of highest-rated product
        """,
        agent=researcher,
    )
    ```
  </Tab>
</Tabs>

<CardGroup cols={2}>
  <Card title="CrewAI Documentation" icon="book" href="https://docs.crewai.com/">
    Dive into the CrewAI documentation to learn more about its capabilities and integrations.
  </Card>

  <Card title="Browserbase Documentation" icon="book" href="https://docs.browserbase.com/">
    Access the Browserbase documentation for comprehensive guides and resources.
  </Card>
</CardGroup>