MetriLLM
Security checks across malware telemetry and agentic risk
Overview
This skill is a straightforward local LLM benchmarking helper, with disclosed local resource use, saved results, and optional public sharing.
Install only if you trust the MetriLLM npm package and source. Expect benchmarks to consume local compute resources and store results locally. Avoid using --share unless you are comfortable publishing model, score, CPU, RAM, and GPU details.
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.