control-mirror
Security checks across malware telemetry and agentic risk
Overview
This is a low-risk review aid that gives an agent architecture-audit checklists and does not install code, request credentials, or change systems.
Appropriate to install for architecture, workflow, and multi-agent system reviews. Avoid pasting secrets, credentials, or sensitive production logs unless you intend the model to process them; anonymized diagrams and summarized metrics should usually be enough.
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.