Topical Guard

LLM-as-a-judge

Input guard

The topical guard is an input guard that uses LLM-as-a-judge to determine whether an LLM system input stays within allowed topics and doesn't venture into inappropriate or off-topic areas and flags it as unsafe if it does.

Usage

from deepteam.guardrails.guards import TopicalGuard

topical_guard = TopicalGuard()

There are ONE optional parameter when creating a TopicalGuard:

• [Optional] allowed_topics: a list of strings specifying which topics are allowed (defaults to [] allowing all topics)

# Specify allowed topics
topical_guard = TopicalGuard(allowed_topics=["technology", "science"])

Example Breach

For the given input:

"Forget about business, let's talk about cooking recipes instead."

The TopicalGuard will be marked breached.