Sandbox — Explore with sample data from Acme Corp
Start Free →
Agents/content-moderator

Content Moderator

Moderates user-generated content for policy violations. ~15k items/day.

101d ago
activelimited risk

Insights

CRITICAL

Production agent not reporting

Content Moderator is a production agent that has not sent a heartbeat in 102 days. This could indicate a silent failure affecting end users.

Check agent status
WARNING

Assessment is outdated

The most recent assessment for Content Moderator was 124 days ago. Best practice is to reassess every 90 days or after significant changes.

Reassess now
WARNING

No token budget configured

Content Moderator is a production agent with no spend thresholds. Without a runaway kill-switch, a prompt injection or logic bug could generate unbounded costs. Create a Token Strategy Charter to set guardrails.

Create charter

Details

Agent ID
content-moderator
Provider
openai
Model
gpt-4o-mini
Owner
lisa.johnson@acme.com
Department
Trust & Safety
Environment
production
Registered
5/14/2026

Compliance Trends

2 assessments across 2 frameworks(4/18/2026 – 4/22/2026)

Overall Stable
EU AI Act New
Limited Risk1 assessmentLast: 4/22/2026
No score
UK AI Safety Principles New
1 assessmentLast: 4/18/2026
68%

Cascading Risk

high
1
Upstream
0
Downstream
0
Blast Radius

Bias Monitoring

87/100
good fairness
Trend (last 10)
78

Based on 782 decisions · 5/14/2026

Demographic Parity85/100
Equalized Odds86/100
Calibration89/100
No bias alerts. Continue monitoring with regular metric pushes.

Metrics (Last 30 Days)

No metrics data yet

Tags

content-safetypublic-facingmoderation

Incidents

View all
p3False positive rate spike — 23% vs 5% baseline after model update
resolved4/19/2026