Security audit
Logical Fallacies
Security checks across malware telemetry and agentic risk
Overview
The reviewed skill artifacts are coherent operational guidance for ClawHub/Convex work, with sensitive actions disclosed and gated by user or admin confirmation.
Install only if you are comfortable giving an agent operational guidance for ClawHub administration, PR maintenance, Convex production work, and staff email flows. Use it with authenticated admin access only when you intend those actions, and keep the documented confirmation and dry-run gates in place.
SkillSpector
By NVIDIA
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
VirusTotal
64/64 vendors flagged this skill as clean.
Static analysis
No suspicious patterns detected.