Docs / Strand / nodes/function-call

Function Call Node

Execute reusable functions from your function library within workflows.

The node references a function by id; editing the function changes every workflow that calls it.

Overview#

Function Call nodes allow you to invoke pre-defined, reusable Python functions stored in your account's function library. This promotes code reuse and keeps workflows clean.

The function library editor under Functions. The reference panel on the right is the authoritative list of what the sandbox allows, and the Test Payload pane runs the function against sample input before you save it.

Available Variables#

Functions use the same secure execution environment as Python Snippets:

Available in Functions#

• payload (dict) - Input data from the previous node
• meta (dict) - Execution metadata
• vault (dict) - Encrypted secrets from the Global Vault (vault['key'] or vault.get('key'))
• return - Return a dict to pass results to the next node

NOT Available in Functions#

• variables - Use Jinja templates
• steps - Use direct connections or Jinja
• context - Not accessible in Python

Configuration#

How It Works#

1. Select a function from your function library
2. The function receives the current event's payload and meta
3. Function code executes in the same secure sandbox as Python Snippet nodes
4. The function's return value becomes the next node's payload

Creating Functions#

Functions are created and managed in the Functions page:

1. Navigate to Functions from the sidebar
2. Click New
3. Give your function a name and description
4. Write your Python code
5. Save the function

Function Code Structure#

Function code follows the same structure as Python Snippet nodes:

# Access input data
user_id = payload.get('user_id')
email = payload.get('email', '')

# Process data
email_normalized = email.lower().strip()
is_valid = '@' in email and '.' in email.split('@')[1] if email else False

# Return result
return {
    'user_id': user_id,
    'email': email_normalized,
    'is_valid': is_valid
}

Available Libraries#

Functions can import the same safe libraries as Python Snippets:

Using Functions in Workflows#

Adding a Function Call Node#

1. Find your function under Functions in the left node palette
2. Drag it onto the canvas
3. The node is configured with that function automatically

Accessing Function Output#

The function's output is available to downstream nodes:

{{ payload.result }}
{{ payload.is_valid }}

Or when accessing from non-directly connected nodes:

{{ steps.function_call_node.output_payload.result }}

Examples#

Email Validation Function#

Function Name: validate_email

Code:

email = payload.get('email', '')

# Email regex pattern
pattern = r'^[a-zA-Z0-9._%+-]+@[a-zA-Z0-9.-]+\.[a-zA-Z]{2,}$'
is_valid = bool(re.match(pattern, email))

# Extract domain
domain = email.split('@')[1] if is_valid else None

return {
    'email': email,
    'is_valid': is_valid,
    'domain': domain
}

Usage in Workflow:

[HTTP Trigger] -> [Function: validate_email] -> [Logic: if_else] -> ...

Data Normalization Function#

Function Name: normalize_user_data

Code:

user = payload.get('user', {})

return {
    'id': user.get('id'),
    'full_name': f"{user.get('first_name', '')} {user.get('last_name', '')}".strip(),
    'email': user.get('email', '').lower(),
    'created_at': meta.get('received_at')
}

Calculate Order Total Function#

Function Name: calculate_order_total

Code:

items = payload.get('items', [])
tax_rate = payload.get('tax_rate', 0.08)

subtotal = sum(
    item.get('price', 0) * item.get('quantity', 1)
    for item in items
)
tax = subtotal * tax_rate
total = subtotal + tax

return {
    'subtotal': round(subtotal, 2),
    'tax': round(tax, 2),
    'total': round(total, 2),
    'item_count': len(items)
}

Function Call vs Python Snippet#

Security#

Function Call nodes execute in the same secure sandbox as Python Snippet nodes, with the same restrictions and resource limits. Vault secrets are accessible via the vault variable, but any vault values in output are automatically replaced with {{ vault.key }} placeholders.

Limitations#

• Same limits as Python Snippets - Execution time and memory constraints
• No external API calls - Use HTTP Request nodes instead
• No file I/O - Use connectors for file operations
• Function must exist - Deleted functions cause workflow failures

Best Practices#

Related#

• Python Snippet Node - Inline Python execution
• Transform Node - Simple data transformations
• Logic Node - Conditional logic and loops

← Back to Strand docs