Security audit

RedHat Code Refactor Engine

Security checks across malware telemetry and agentic risk

Overview

This is a text-only refactoring guide that asks the agent to change code only in the normal course of user-directed refactoring, with tests and validation as safeguards.

Reasonable to install for agent-guided refactoring. Review proposed code changes before merging, make sure tests or characterization checks exist, and consider asking the publisher to fix the malformed trigger metadata. The RedHat-style naming is not verified by these artifacts, so do not assume official affiliation from the package text alone.

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: 95% confidence
Finding: The trigger configuration is malformed or underspecified, using a generic serialized type marker (`System.Object[]`) instead of concrete activation phrases or structured constraints. This can cause the skill to be invoked unpredictably or too broadly, increasing the chance of unintended activation and unsafe execution in contexts where the user did not explicitly request refactoring.

VirusTotal

63/63 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.