← Glossary / 70 of 100
n. / Agents and tools
Guardrails
Definition
Guardrails are checks placed around a model or agent that block, change or flag inputs and outputs that break your rules, such as unsafe content, personal data, off-topic requests or invalid tool arguments.
In more detail
They can be simple code, like a schema check or an allow-list of tools, or a second model that classifies content. They work best as layers: validate inputs, restrict what tools can do, and check outputs. Log every time a guardrail fires, since the count and the reasons are a useful quality signal.
Example
Before replying, a check scans the draft for card numbers and email addresses. If it finds any, the reply is blocked and the event is recorded on the run.
