Security

fluiq.secure() adds a scanning layer to every traced LLM call. Choose warn to flag risks on the trace, or block to halt the request before it reaches the model.

Default mode — every traced LLM call is scanned server-side. Risks are flagged on the trace without blocking execution.

Python

import fluiq
import openai

fluiq.instrument(api_key="fl_...")
fluiq.secure()  # warn mode by default

client = openai.OpenAI()

# Calls proceed normally; risk metadata appears in the Security dashboard
response = client.chat.completions.create(
    model="gpt-4o",
    messages=[{"role": "user", "content": user_input}],
)