Policy providers

Providers evaluate a specific property of submitted text or tool telemetry. Read each provider’s scope before enabling it. Model-backed providers require their configured models or judge endpoints; a local installation does not automatically have all detectors ready.

Data and application boundaries

• Personal data (PII) — pii_detection
• Prompt injection — prompt_injection
• Jailbreak attempts — jailbreak
• Toxicity — toxicity
• Bias signals — bias_detection
• NSFW text — nsfw_detection
• Grounded output — hallucination
• Gibberish heuristics — gibberish_detection
• Secrets detection — secrets_detection
• SQL injection patterns — sql_injection
• Unsafe web content — web_sanitization
• Topic restrictions — topic_restriction
• Competitor mentions — competitor_check
• Sensitive topics — sensitive_topics
• Unusual prompt patterns — unusual_prompt
• Tool authorization — tool_authorization

Start with one measurable check

Use Secrets detection and the quickstart for a deterministic synthetic test. Add one provider at a time and compare safe, unsafe, ambiguous and unavailable cases before changing enforcement.

Policy configuration belongs to the saved policy. Passing policy types through the evaluation request selects ad-hoc checks with default configuration; it does not load the saved configuration of a policy ID. See the API reference.

Health, completed evaluation, measured accuracy and application enforcement are separate acceptance criteria. Use the policy workflow and support matrix to record them.