SightShield

StableContent SafetyPARROTSIGHT native

Multilingual content safety for regulated platforms

P50 95msP95 180ms$0.35 / 1k calls

Overview

SightShield classifies user-generated and enterprise content against 14 safety categories with per-label confidence scores. It is calibrated for Malaysian regulatory context (PDPA, Communications and Multimedia Act) and supports custom threshold policies per category.

14 moderation categories with per-label scores
Policy thresholds configurable per category
EN / MS / ZH native support
PII detection (MyKad, passport, card numbers)

Capabilities

Category accuracy95

MS language quality90

ZH language quality92

Throughput headroom97

API

https://api.parrotsight.com/v1/moderation

bash

curl -X POST https://api.parrotsight.com/v1/moderation \
  -H "Authorization: Bearer $PS_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"ps-shield-base","input":"Moderate a user comment across violence, hate speech and PII.","threshold":0.8}'

Version history

v1.8.3Latest2026-08-14
MyKad number pattern v3 support
v1.8.0Released2026-06-02
14th category: self-harm discourse
v1.7.0Released2026-03-19
Policy threshold API GA

More models

View all models

ps-guard-maxStable

SightGuard Max

Threat intelligence and anomaly detection at network scale

$2.80 / 1k calls

ps-guard-maxStable

SightGuard Max

Threat intelligence and anomaly detection at network scale

$2.80 / 1k calls