Security audit
Red Note Content Topic
Security checks across malware telemetry and agentic risk
Overview
The artifacts describe operational ClawHub, Axiom, Convex, and documentation skills with powerful but disclosed, task-directed behavior.
Install only if you intend to use these operational skills with the relevant service accounts. Review which tokens and admin roles are configured locally, prefer least-privilege API tokens, and be careful with commands that send email, change moderation state, delete or recreate monitors/dashboards, deploy production, or run migrations.
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.