Error Prone

• Outcome: Pass
• Latest audit: May 29, 2026 at 8:26 PM UTC
• Version:
  1.0.0

Overview

This is an instruction-only writing style skill that openly adds human-like imperfections and does not request access to files, accounts, credentials, network services, or persistent execution.

Install only if you want an opt-in style transformation that intentionally adds imperfections. Avoid using it where AI-use disclosure, academic integrity, professional records, legal or business accuracy, or recipient trust matters.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Trigger Abuse: Overly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
• 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
• Supply Chain: Unpinned Dependencies, External Script Fetching, Obfuscated Code

Findings (1)

Vague Triggers

Medium

• Confidence: 96% confidence
• Finding: The trigger description is overly broad and can activate on common writing requests such as 'humanize this' or 'make it less AI,' which are ordinary user intents rather than narrowly scoped commands. That increases the chance of unintended invocation, causing the agent to silently alter outputs in ways the user did not explicitly request and potentially degrading reliability or policy compliance in professional contexts.

VirusTotal

64/64 vendors flagged this skill as clean.