Security audit
content-sanitization
Security checks across malware telemetry and agentic risk
Overview
This skill is a safety-focused markdown guideline for sanitizing untrusted content and does not include executable code or hidden installation behavior.
This skill is reasonable to install as guidance for agents that process external content. Treat its automated-hook language as informational unless the separate plugin hook is actually installed, and continue to review any workflow that fetches public web or GitHub content.
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.