Prompt defense
Security checks across malware telemetry and agentic risk
Overview
This is a defensive, instruction-only email safety skill that contains prompt-injection examples so an agent can detect and block them.
Reasonable to install as a defensive aid for email workflows. Be aware that the documentation intentionally contains adversarial prompt examples and hidden-character samples; treat them only as signatures to detect, and keep the user-confirmation step enabled before taking any action requested inside an email.
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
60/60 vendors flagged this skill as clean.