Step: Guardrail

• Key: guardrail
• Category: AI & Language Models
• Description: Validate content against specified criteria to ensure compliance.

Inputs

• modelId (select, required): Model ID
• prompt (long-text, required): Content to validate
• restrictedTopics (long-text): Comma-separated topics that must not appear
• customInstructions (long-text): Additional validation instructions

Outputs

• isValid (boolean): Whether content passed validation
• message (string): Explanation or failure reason

Notes

• Place before sending user content to LLMs or external systems.