> For the complete documentation index, see [llms.txt](https://documentation.maax.ai/main/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://documentation.maax.ai/main/setup-your-integration/v2-sales-ai-zapier/1a.-explanation-of-python-code-step.md).

# 1a. Explanation of Python Code Step

{% embed url="<https://www.loom.com/share/60283bc5fc984444840ba7dbf00b64d2?sid=eb35b905-0e0b-4567-b7b4-ffaaab27dbe0>" %}

{% code title="Sample Python Code" %}

```python

import requests

api_key = # 'input your API key here'

url = 'https://app.maax.ai/api/v2/outgoing-messages/'
headers = {
    '2hd-api-key': api_key
}

service_type = # 'customer_support_automation' or 'sales_automation' 
platform = # 'email, 'website, or 'sms'
ai_id = # 'input your ai id here'
followup_url = # 'input your followup webhook url here'
send_response_url = # 'input your followup webhook url here'

message_content = input_data['message']
phone_number = input_data.get('phone', '') 
first_name = input_data.get('first_name', 'unknown')

data = {
    'service_type': service_type,
    'platform': platform,
    'ai_id': ai_id,
    'followup_url': followup_url,
    'send_response_url': send_response_url,
    'message_content': message_content,
    'prospect_info': {
        'phone': phone_number,
        'first_name': first_name
    }
}

# POST request
try:
    response = requests.post(url, json=data, headers=headers)
    response.raise_for_status()  
    output = [response.json()]

except requests.exceptions.RequestException as e:
    error_message = str(e)
    output = {"error": error_message}
```

{% endcode %}

## Summary

1. **`Simplifying the Code Step`** Despite its technical appearance, the code step in Zapier has been simplified to mainly involve copy-pasting and filling in different values.
2. **`Utilizing 'Code by Zapier' Step`** This step is used after everything in the 'ingest' part of the process. It involves running Python code.
3. **`Guidance on Usage`** The speaker encourages pausing and revisiting sections of the tutorial as needed, emphasizing the ease of the process.
4. **`Mapping Input Data`** The tutorial focuses on mapping specific fields like message, phone, and first name in the input data to the code.
5. **`Extracting User Query Information`** The 'message' field involves dropping in the user's question, which will be sent to Maax AI for processing.
6. **`Additional Data Fields`** Other data fields like phone and first name are also sent over. The user has the option to add more information like email and Instagram based on their needs.
7. **`API Key Handling`** The API key is discussed, with a recommendation to use 'storage by Zap' in Zapier for better security, instead of directly showing API keys.
8. **`Setting up Sales Automation and AI ID`** The user selects their sales automation platform and inputs the AI ID from their dashboard, which links the specific version of AI to the Zapier step.
9. **`Establishing Personalized Responses`** Different AI IDs may represent different use cases, each driving a unique goal, thus establishing different 'personalities' for the responses.
10. **`Setting Up Follow-Up URLs`** The tutorial includes mapping webhooks from Zapier in the body response, which is crucial for sending responses to Zapier triggers.
11. **`Finalizing the Step and Testing`** After setting up, a test is run to ensure that the output is successful and that the data is correctly sent to Maax AI.

<br>