Security audit
My Clean Skill
Security checks for vulnerabilities and agentic risk
Overview
The skill artifacts are coherent with developer, moderation, PR-review, UI-proof, and Convex setup workflows, with no evidence of hidden exfiltration or destructive behavior.
Install only if you expect these skills to operate on a ClawHub/Convex development environment. Review the moderation and autoreview workflows carefully: moderation commands can change public/account state, and autoreview's helper defaults to a full-access nested review unless run with --no-yolo or AUTOREVIEW_YOLO=0.
Vulnerability Patterns
- Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
- Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
- Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
- Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
- Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Static analysis
No suspicious patterns detected.