Qa Gate

• Outcome: Review
• Latest audit: May 29, 2026 at 9:39 PM UTC
• Version:
  1.0.0

Overview

This QA skill is a normal review checklist, but it needs review because it claims to be read-only while also telling the agent to edit artifacts and create report files.

Install only if you are comfortable with a QA workflow that may create report files and direct the agent to edit documents or code after failures. For safer use, ask it to report findings first, require explicit approval before fixes, and avoid dual cross-model validation for sensitive content unless you understand where the content will be sent.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Trigger Abuse: Overly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
• MCP Tool Poisoning: Hidden Instructions, Unicode Deception, Parameter Description Injection
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Privilege Escalation: Excessive Permissions, Sudo/Root Execution, Credential Access

Findings (5)

Intent-Code Divergence

Medium

• Confidence: 98% confidence
• Finding: The skill says it only inspects artifacts, but Step 4 directs the agent to fix failed artifacts and re-validate them. That turns a supposedly passive review skill into one that may alter user documents or code without clear authorization, increasing the risk of unintended or unsafe changes.

Intent-Code Divergence

High

• Confidence: 98% confidence
• Finding: The skill says it only inspects artifacts, but Step 4 directs the agent to fix failed artifacts and re-validate them. That turns a supposedly passive review skill into one that may alter user documents or code without clear authorization, increasing the risk of unintended or unsafe changes.

Vague Triggers

Medium

• Confidence: 92% confidence
• Finding: The trigger phrases are broad enough to match common requests like 'proofread this' or 'final check,' which can cause the skill to activate in many ordinary contexts without clear user intent. Over-broad invocation increases the chance of unexpected behavior, especially because the skill includes file-writing and possible artifact modification steps.

Vague Triggers

Medium

• Confidence: 90% confidence
• Finding: The usage guidance says the gate should run on nearly any artifact before review, making the scope overly expansive. In practice, this can cause the skill to be auto-selected for a wide range of tasks where users did not intend a potentially state-changing QA workflow.

Missing User Warnings

Medium

• Confidence: 97% confidence
• Finding: The skill instructs the agent to write a validation report into the workspace but does not warn the user that files will be created. Silent filesystem writes violate least surprise and can be abused to leave artifacts, overwrite expected paths, or create compliance issues in read-only or sensitive environments.

VirusTotal

64/64 vendors flagged this skill as clean.