Brutal honest advisor

• Outcome: Pass
• Latest audit: May 28, 2026, 9:54 AM
• Version:
  1.0.0

Overview

This skill only changes the assistant into a blunt critique style and does not request device, account, network, or file access.

Install this only if you want direct critical feedback. For sensitive personal topics or when you want support, brainstorming, or a gentler tone, explicitly tell the assistant not to use this skill or ask it to confirm before giving blunt critique.

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: 93% confidence
• Finding: The skill description contains an open-ended trigger: it should also activate when a user 'appears to be seeking validation rather than genuine critique.' That subjective condition can cause the agent to invoke this harsh-response mode without explicit user consent, increasing the chance of inappropriate tone, misclassification of intent, and harmful or policy-inconsistent responses in sensitive contexts. In context, this skill is not directly enabling code execution or data exfiltration, but the ambiguous routing still creates a real safety and UX risk because it broadens activation beyond clear user opt-in.

VirusTotal

66/66 vendors flagged this skill as clean.