code-review-skill

Security checks across malware telemetry and agentic risk

Overview

This is a straightforward code-review skill that may produce fixes and tests, with no evidence of hidden installation, persistence, credential use, or data exfiltration.

Install this only if you want an agent to inspect staged code changes and suggest or generate fixes and tests. Use it on repositories and diffs you intend to share with the agent, and review any generated changes before applying or committing them.

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
94% confidence
Finding
The skill’s trigger/request wording is extremely broad: it asks for a comprehensive review of staged changes in terms that overlap with many normal developer requests. That can cause over-invocation or invocation in unintended contexts, exposing unrelated code or causing the agent to perform expansive review and code-generation actions without clear scope boundaries.

VirusTotal

63/63 vendors flagged this skill as clean.

View on VirusTotal