Security audit
Devops Scripts
Security checks across malware telemetry and agentic risk
Overview
The skill bundle is mostly coherent, but needs review because its autoreview helper gives a nested agent full local access by default.
Install only in a trusted ClawHub maintainer or development environment. Before using the autoreview helper, prefer --no-yolo or AUTOREVIEW_YOLO=0 unless you intentionally want a nested agent to have full local access. Treat moderation and proof-publishing commands as real external mutations that can affect users, roles, tokens, public comments, and published artifacts.
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.