​
Python Script Action

Use the core.script.run_python action to execute Python code in a sandboxed WebAssembly environment using Pyodide. This ensures isolation from the host system while providing access to popular Python libraries. You can include 3rd-party Python packages in your script by adding them to the dependencies field.

  • Scripts must contain at least one function
  • If multiple functions exist, one must be named main
  • The function’s return value becomes the action output
  • The output must be JSON serializable (e.g. str, int, float, bool, list, dict)
  • Network access is disabled by default for security; enable with allow_network: true when using dependencies

​
Configuration

FieldTypeDescriptionRequired
scriptstringPython code with at least one function✅
inputsobjectDictionary mapping parameter names to values for function arguments❌
dependenciesarrayList of Python packages to install❌
timeout_secondsintegerMaximum execution time (default: 30)❌
allow_networkbooleanEnable network access for package downloads (default: false)❌
  • inputs are passed as function arguments into the main function
  • Keys must match the parameter names in the function signature
  • Missing parameters will receive None
  • Functions must have parameters to receive inputs
  • Extra input fields that don’t match function parameters are ignored
  • If your function has def main(name, age): but inputs include {"name": "Alice", "age": 30, "city": "NYC"}, the city field will be ignored without error.

​
Examples

The following examples show valid inputs for the Run Python script action.

script: |
  def main():
    # Simple arithmetic
    result = 10 * 5 + 3
    return result

script: |
  def main(a, b):
    # Simple arithmetic
    result = a * b + 3
    return result
inputs:
  a: 10
  b: 5

script: |
  def calculate_tax(amount, rate):
    return amount * rate

  def add_fixed_tax(amount, add_on=10):
    return amount + add_on

  def main(subtotal, tax_rate):
    # When multiple functions exist, 'main' is called
    tax = calculate_tax(subtotal, tax_rate)
    total = add_fixed_tax(subtotal, 10)
    return total + tax
inputs:
  subtotal: 100.0
  tax_rate: 0.08

script: |
  def main():
    import numpy as np
    result = np.array([1, 2, 3])
    return result.tolist()
dependencies:
  - numpy
allow_network: true

script: |
  def main(user_name: str, age: int, city: str, country: str = "USA"):
    if country is None:
      return f"{user_name} ({age}) lives in {city}"
    else:
      return f"{user_name} ({age}) lives in {city}, {country}"
inputs:
  user_name: "Alice"
  age: 30
  city: "New York"