Security audit

Skills

Security checks for vulnerabilities and agentic risk

Overview

This skill is a straightforward writing-style analysis and imitation helper with no hidden execution, persistence, credential use, or unrelated data access.

Installers should understand that this skill may activate for many writing-style or imitation requests. Use it with content you have rights to analyze, avoid deceptive impersonation, and do not use it to reproduce copyrighted text verbatim.

SkillSpector

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 (2)

Vague Triggers

Medium

Confidence: 91% confidence
Finding: The skill declares trigger conditions so broadly that it can activate on many ordinary requests about analyzing tone, style, or imitating writing, even when a dedicated skill is not clearly intended. Over-broad routing can cause unintended capability invocation, bypass normal task scoping, and increase the chance of assisting with sensitive style imitation or copyrighted-text transformation without sufficient checks.

Vague Triggers

Medium

Confidence: 88% confidence
Finding: The natural-language examples are expansive and lack exclusions, making this skill likely to capture general writing-assistance requests that should remain in the default assistant flow. In security terms, ambiguous dispatch increases the attack surface: users can trigger specialized behavior through loose phrasing, including style replication workflows that may raise impersonation, abuse, or copyright-adjacent concerns.

Static analysis

No suspicious patterns detected.