Quick Start

Get SentinelAI running in under 60 seconds.

1. Install

pip install sentinelai-risk

2. Verify a response

from sentinelai import SentinelAIClient

client = SentinelAIClient(api_key="sk_...")

prompt = "What was Q3 revenue?"
llm_output = "Revenue grew 45% year over year."

result = client.verify(prompt=prompt, response=llm_output)

if result["status"] == "hallucinated":
    print(result["corrected"])  # serve the fix, not the flaw
else:
    print(result["score"], result["claims"])

That's it. Every response is now scored, explained, and protected.

3. What you get back

verify() returns a dict with:

• score — a number from 0 (safe) to 100 (critical risk)
• status — trusted (0–24), needs_review (25–59), or hallucinated (60–100)
• claims — detector-level findings with severity and note
• corrected — the cleaned response, present when correction applies
• decision / action_taken — what the policy engine decided and did

Next steps