Guards#
Every guard is a method on an Enkrypt object and returns a guard ready for the harness guardrails.
Guard |
Checks for |
Default action |
|---|---|---|
prompt injection and jailbreak attempts |
|
|
hateful, harassing, violent or otherwise toxic text |
|
|
your own rules, inline or saved in Enkrypt |
|
|
API keys, passwords, tokens |
|
|
personal data of the types you list |
|
|
whatever a guardrail saved in Enkrypt checks |
per detector |
Every guard takes:
on_violation: what happens when Enkrypt flags the value; see Actions;on_error: what happens when Enkrypt cannot be reached,"block"or"allow"; see Errors and enforcement.