Nervtimer
Security checks across malware telemetry and agentic risk
Overview
The skill bundle is mostly coherent and disclosed, but its review helper defaults to sandbox-bypassing full access and can send diffs to external reviewer tools.
Review before installing in a sensitive workspace. Use the moderation and PR skills only with the intended ClawHub/GitHub accounts, and consider disabling autoreview full-access mode or external fallbacks (`--no-yolo`, `AUTOREVIEW_YOLO=0`, or fallback reviewer `none`) unless you explicitly want those behaviors.
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
65/65 vendors flagged this skill as clean.