Security audit
ia-refine-prompt
Security checks across malware telemetry and agentic risk
Overview
This skill guides an agent to rewrite prompts more clearly and only offers an optional local save after user confirmation.
Review the refined prompt before using it, especially for system prompts or agent instructions. The optional save behavior is disclosed and should only occur after you approve it.
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.