Engineer

Security checks across malware telemetry and agentic risk

Overview

The available evidence shows a purpose-aligned engineering helper with only a review note about broad activation, not malicious behavior.

Install only if you want an engineering-advice mode that may activate broadly for technical tasks. Review its activation wording and memory behavior in your client settings, and disable or narrow it if you only want it used on explicit engineering-review requests.

SkillSpector

By NVIDIA

Vulnerability Patterns

Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
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

Findings (1)

Vague Triggers

Medium

Confidence: 90% confidence
Finding: The activation criteria are broad enough to match many normal technical requests, which can cause the Engineer skill to activate more often than the user explicitly intended. In a system with persistent memory behavior, over-broad activation increases the chance of unneeded context capture, unintended steering of conversations, or inappropriate application of this skill's behavioral rules across unrelated sessions.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal