Criticism Self Criticism

• Outcome: Pass
• Latest audit: May 29, 2026 at 8:32 PM UTC
• Version:
  1.0.0

Overview

This is a markdown-only self-review skill with broad review triggers but no code execution, data access, persistence, or hidden behavior.

Install this only if you want an explicitly Chinese-language, Maoist-framed self-review methodology to influence review and retrospective workflows. It appears technically low risk, but teams should decide whether the framing and broad activation language fit their norms.

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: 95% confidence
• Finding: The trigger conditions are very broad and can match many routine completion, review, or feedback situations, causing the skill to activate far outside narrowly intended use. In an agent system, this can steer behavior frequently and disproportionately, especially because the skill promotes a specific ideological review framework that may override more task-appropriate guidance.

VirusTotal

65/65 vendors flagged this skill as clean.