Software Architecture Design SOP

Security checks across malware telemetry and agentic risk

Overview

This is a documentation-only architecture-design skill, with one privacy caution around ML logging guidance but no evidence of hidden execution or data access.

Safe to install for architecture documentation. When using the ML/AI section, avoid blindly logging raw prompts, inputs, outputs, uploaded documents, or personal data; prefer metadata, sampling, redaction, retention limits, encryption, and restricted log access.

SkillSpector

By NVIDIA

Vulnerability Patterns

Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
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

Findings (1)

Missing User Warnings

Medium

Confidence: 93% confidence
Finding: The guidance explicitly says to log model inputs and outputs for drift monitoring, but provides no guardrails for sensitive data minimization, redaction, consent, retention, or access control. In an ML/AI system context, inference payloads often contain PII, confidential business data, prompts, uploaded documents, or model outputs derived from sensitive inputs, so following this advice naively can create a privacy and data-exposure risk.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal