Experiment Lifecycle Governance
Security checks across malware telemetry and agentic risk
Overview
This skill is not malware, but it needs review because it weakens its own safety guard for destructive experiment actions by encouraging PIN bypass in automation.
Review before installing. The metrics and audit guidance are purpose-aligned, but use the destructive-operation controls carefully: avoid command-line PINs, do not normalize `--force`, and require scoped automation permissions plus audit logging before canceling, stopping, or deleting experiments.
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.