Safe Exec Wrapper
Security checks across malware telemetry and agentic risk
Overview
This is a small defensive wrapper that labels command output as untrusted, but it does not sandbox or make the wrapped command itself safe.
Install this if you want an agent-facing helper that marks untrusted command output with boundaries. Keep reviewing commands before passing them to safe-exec, especially commands that modify files, contact services, or use credentials, because the wrapper does not sandbox execution or prevent side effects. Use the default random UUID mode for normal use.
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.