Security audit
Finalize Agent Prompt
Security checks across malware telemetry and agentic risk
Overview
This is a simple prompt-polishing skill that only asks the agent to edit a user-provided prompt file.
Only provide prompt files you want the agent to revise. Static scan and SkillSpector reported no issues, and the pending VirusTotal telemetry does not change the verdict without artifact-backed risk.
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.