Llm Eval Router
Security checks across malware telemetry and agentic risk
Overview
This skill openly describes an LLM evaluation router that uses your own provider API keys and optional scheduled runs; no hidden or malicious behavior was found.
Install only if you are comfortable sending evaluated prompts and outputs to the named LLM providers using your own API keys. Enable the suggested cron/systemd/launchd schedule only deliberately, keep provider keys scoped and out of logs, and avoid routing high-safety or highly confidential tasks through this pipeline unless your provider and data-retention settings are acceptable.
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.