Refactoring
Security checks across malware telemetry and agentic risk
Overview
This is an instruction-only refactoring workflow with no executable code, hidden behavior, credential access, or persistence.
Safe to install as a refactoring checklist. Users should still review any code changes the agent makes, keep refactoring separate from feature work when possible, and ensure tests or manual checks match the risk of the touched 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
64/64 vendors flagged this skill as clean.