Security audit
Social Science Journal Abstract Polisher
Security checks across malware telemetry and agentic risk
Overview
This is a prompt-only academic editing skill with no code, installer, credential use, persistence, or external data access.
This skill appears appropriate for polishing abstracts, but users should review the final text to ensure it does not change research claims or overstate results. For sensitive or unpublished research, only use it in an agent environment you trust with that text.
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.