Custom Script

---

Purpose

The CustomScript node allows you to execute custom code inside ThreoAI workflows.
It supports Python, enabling complex data processing, calculations, integrations, and business logic beyond standard nodes.

📥 Inputs

• Data from previous workflow nodes is passed into the script.

📤 Outputs

• Script output is passed to the next workflow node.
• Must be wrapped inside a unique JSON object name (e.g., {"custom_result": {...}}).

Output Format (example): json { "custom_result": { "processed": true, "score": 95, "status": "approved" } }

⚙️ Parameters

Name	Type	Required	Default	Description
Script Content	Code editor	✅	Empty	Python/JavaScript code executed by this node.
Programming Language (language)	Dropdown	✅	Python	Defautl to Python.
Script Encryption (encryption)	Toggle	Optional	Off	Protects script code with encryption.
Encryption Password (password)	Password	Required if encryption enabled	Empty	Password required to view/edit encrypted scripts.

💡 Example Usage

• Loyalty Points Calculator (Python): Apply custom rules combining purchase data, customer tiers, and seasonal bonuses.
• Invoice Validation: Flag incomplete or suspicious invoice data before payment processing.
• Sales Analytics: Calculate growth rates, quotas, and rankings across multiple data sources.

📘 Best Practices

• Start with simple scripts and test often.
• Apply error handling (try/except) to avoid workflow failures.
• Always return results wrapped in a unique JSON object name.

⚡ Python Execution Contract

Each Python script must define an execute function as the entry point:

    def execute(input_json, meta):
        # input_json: data received from the previous node
        # meta: metadata provided by the workflow engine

        # Example: echo input with added field
        result = {
            "custom_result": {
                "original": input_json,
                "processed": True
            }
        }

        return result
```text
### Rules

- `input_json`: Contains data passed from the previous node.
- `meta`: Contains workflow context and metadata.
- Output must be wrapped in a **unique JSON object name** (e.g., `"custom_result"`) to avoid collisions with other nodes.
- The return value must be JSON-serializable.