Security audit
Agent Directory
Security checks across malware telemetry and agentic risk
Overview
This is a simple directory skill that fetches public service listings and linked skill files, with no credential use, persistence, or hidden execution.
This skill is reasonable to install as a public directory. Treat any linked service skill.md as a new, untrusted instruction source: review its requested commands, credentials, data access, and account actions before allowing the agent to follow 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
65/65 vendors flagged this skill as clean.
Static analysis
No suspicious patterns detected.