Security audit
Ai Agent Configurator
Security checks across malware telemetry and agentic risk
Overview
The skill bundle is coherent for ClawHub maintainer and Convex development work, but it includes high-impact moderation/GitHub authority and a review helper that defaults to full-access nested Codex execution.
Install only in a trusted ClawHub maintainer or developer environment. Use the moderation confirmation and audit-log safeguards, keep GitHub and ClawHub tokens narrowly scoped, and prefer the autoreview helper's --no-yolo option unless you intentionally want nested Codex review with full filesystem and command authority.
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.