Check text for profanity and get detailed results including matched words, severity, and positions

Censor profanity in text by replacing matched words with a replacement character

Analyze text with context awareness - considers domain-specific whitelists

Check multiple texts for profanity in a single operation (max 100)

Get the list of all 24 supported languages

Validate content with a safety score (0-100) and action recommendation

Detect text obfuscation techniques like leetspeak and Unicode homoglyphs

Explain why a word was flagged with detailed reasoning

Suggest clean alternatives for profane content

Analyze a collection of texts for profanity statistics (max 500)

Compare detection results across different strictness levels

Generate regex patterns for custom profanity detection

Track a user message and update their risk profile - enables pattern detection across messages

Get a user's moderation profile and risk assessment

Get list of users with high risk scores for review

Reset a user's moderation profile (use after appeals or timeouts)

Add message to real-time processing queue and get instant result - optimized for chat streams

Process multiple messages in real-time batch mode - for high-throughput scenarios

Get real-time streaming statistics and performance metrics

Scan text for prompt injection attacks using rule-based pattern matching. Returns a risk score (0-1), a decision (ALLOW / HITL / BLOCK), matched categories, and per-match position details. Custom patterns: max 20 patterns, each regex max 200 chars, 50 ms execution budget per pattern.

Scan text for leaked credentials, API keys, tokens, and other secrets. Returns a full ScanResult with per-match details including pattern id, family, and character positions.

Scan text for Personally Identifiable Information (email, phone, SSN, credit card, IBAN, IP, MAC, passport, date of birth, etc). Returns a full ScanResult with per-match details. Use redact_pii for a reversible vault-backed redaction.

Redact PII from text using a server-side vault for reversible round-trip. Returns sanitized text with [REDACTED__N] placeholders, a vaultId to use for restoration, and a list of placeholder+type pairs. Original values never leave the server. Call restore_pii with the same vaultId to get originals back.

Restore PII placeholders in text back to their original values using a vault session created by redact_pii. The vault is maintained server-side — originals were never sent to the AI client.