Analyze text or images for potentially harmful content.

Prompt Shields is a unified API that analyzes LLM inputs and detects adversarial user input attacks.

The invisible chars rule detects and blocks the prompt if it contains any Unicode or similar hidden characters.

The "Prompt Injection LLM Rule" detects and mitigates prompt injection attacks within the input, ensuring the security and integrity of the system.

The "PII Filter" rule detects, anonymizes, and removes any Personally Identifiable Information (PII) from a prompt, ensuring privacy and data protection.

The "Detect English" rule enforces the exclusive use of English within the prompt, restricting any inclusion of other languages.