Now in beta
Skill or
Scam?
Check your agent skills before the agent is turned against you.
40+ pattern checks and AI semantic analysis for prompt injection, data exfiltration, credential harvesting, and hidden content.
⚡FREE
Static Analysis
40+ patterns catch prompt injection, exfiltration, obfuscation, and hidden content instantly — runs in your browser.
🤖3 FREE / MO
AI Deep Scan
Claude Sonnet performs semantic analysis to catch threats that regex can't — subtle manipulation, scope violations, social engineering.
🔗PRO
CI/CD Integration
Webhook API scans skills automatically on push. Block malicious skills before they reach production.
Detects 7 threat categories
🧠 Prompt Injection📡 Data Exfiltration📂 File System Abuse🔮 Code Obfuscation🔑 Credential Harvesting👻 Hidden Content🚧 Scope Violation