Security audit
Aa Benchmarking Framework
Security checks across malware telemetry and agentic risk
Overview
This is a draft, instruction-only benchmarking skill with no executable code, credentials, network access, or hidden install behavior.
This appears safe to install as a draft benchmarking guide. Re-review before relying on future versions that add executable code, LangFuse ingestion, credential handling, or memory-backed behavior.
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.