← Back to Glossary
Guardrails
Phase 04 · Agentstechnique

Input and output validation layers that enforce safety rules, topic boundaries, and formatting constraints.

Why it exists

Prevents prompt injection attacks, PII leaks, inappropriate language, and malformed tool parameters before execution.

How it works

Runs classifier models, regex filters, or Pydantic validators on inputs and outputs before passing them onward.

Where you'll use this