Classement document
Security checks across malware telemetry and agentic risk
Overview
The skills are coherent development and ClawHub maintenance helpers, with sensitive actions disclosed and guarded by user-directed commands.
Install this only if you want agent help with Convex and ClawHub maintainer workflows. Review before using moderation commands because they can affect real users and public skill visibility, and consider running autoreview with --no-yolo or disabling fallback reviewers when working with sensitive private code.
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.