Security audit

Quality Writer Pro

Security checks for vulnerabilities and agentic risk

Overview

This is a plain writing-workflow skill with no code, dependencies, persistence, or hidden execution, though users should approve any publishing step before it is performed.

Install this if you want a structured writing and quality-review workflow. Before using it with any blog, CMS, social media, or connected publishing account, require the agent to show the final content, destination, and publish action for approval; also tell it what language to use if you do not want Chinese.

SkillSpector

Vulnerability Patterns

Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
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: 90% confidence
Finding: The skill’s invocation guidance is broad enough that an agent may apply it in many writing-related contexts without clear boundaries, increasing the chance it overrides higher-priority user intent or is invoked when unnecessary. In agent systems, vague activation criteria can create prompt-scope creep, where a generic workflow starts steering behavior beyond the user’s explicit request.

Natural-Language Policy Violations

Medium

Confidence: 86% confidence
Finding: The metadata states Chinese as the applicable language without clearly making that conditional on user preference, which can cause the agent to respond in Chinese even when the user asked for another language. This is primarily an instruction-governance issue: it can degrade usability, create confusion, and indirectly override user intent, though it is not a direct security compromise.

Static analysis

No suspicious patterns detected.