llm-benchmark-analyst
Security checks across malware telemetry and agentic risk
Overview
This is an instruction-only benchmark research skill that asks the agent to browse public benchmark sources and produce cited reports, with no code execution, credential use, persistence, or hidden data handling found.
Before installing, understand that the skill will likely browse external public leaderboard sites and may read image-rendered tables to answer benchmark questions. Verify cited scores for important decisions, especially because the package has limited publisher provenance, but the reviewed artifacts are coherent and purpose-aligned.
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
66/66 vendors flagged this skill as clean.