🛡️ PolicyGuard-RL
Multi-Agent Content Moderation Environment
Ready
Misinformation (Easy)
Hate Speech (Medium)
Harassment (Medium)
Child Safety (Hard)
Violence (Hard)
Start Episode
Reset
Escalation Chain
Tier 1
T1
First-Line
→
T2
Senior
→
T3
Appeals
Content to Review
--
Click "Start Episode" to begin reviewing content.
👤
--
❤️
0
🔄
0
⚠️
0
✓ Allow
✗ Remove
↓ Reduce
⚑ Label
↑ Escalate
Metrics
0%
Accuracy
0%
Escalation
0.00
Score
0
Steps
Activity Log
System ready. Select a task and start an episode.