Security audit
Generate Judgements
Security checks across malware telemetry and agentic risk
Overview
This Markdown-only skill helps generate evaluation YAML and only writes files after user review and approval.
Install this when you want help generating mlflow-skills judge_definitions. Review the proposed scopes, generated judgements, and exact YAML path before approving writes, and keep real API keys in environment variables or a secret manager rather than in generated config files.
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.