Noqoro | AI Policy Controls

Content Filtering

Self Harm Detection

Redact outputs containing self-harm instructions.

Hate Speech Restriction

Threshold for restricting toxic outputs.

Strictness Strict

Violence Restriction

Filter parameters for graphic content.

Strictness Medium

PII Detection

Social Security Numbers (SSN)

Scan for and redact 9-digit SSN format in LLM payloads.

Credit Card Credentials

Redact PCI-DSS credit card numbers automatically.

Phone Numbers

Filter national and international telephone sequences.

Custom Regex Redaction

Apply custom regular expression matches.

Operational Guardrails

Prompt Injection Protection

Inspect prompt payload for hostile system override commands.

Sensitivity

Model Leak Detection

Prevent output of raw system parameters or files.

Sensitivity

Model Behavior

Sentiment Analysis

Restrict system expressions based on target sentiment.

Allowed

Tone Enforcement

Verify system messages adhere to corporate persona rules.

Required Tone

Accuracy & Factuality Check

Enforce strict grounding comparison against uploaded datasets.

Threshold

Policy Controls