Security audit

中文文本校验

Security checks for vulnerabilities and agentic risk

Overview

This is a straightforward Chinese document proofreading skill with no code, network behavior, persistence, or hidden high-impact actions.

Install this if you want Chinese text proofreading for documents. Be aware that broad trigger words like proofreading or checking documents may activate it for some general editing tasks, and only provide document files whose contents you are comfortable having the agent read for proofreading.

A.I.G

Vulnerability Patterns

Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs

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 (3)

Natural-Language Policy Violations

Medium

Confidence: 85% confidence
Finding: L004-L007 与全文均将技能限定为中文文本校验，但未说明这是用户自选的语言模式，也未提供其他语言的选择或显式 opt-in。按规则，强制特定语言/locale 而无用户选择可构成自然语言层面的语言政策问题。

Vague Triggers

Medium

Confidence: 97% confidence
Finding: The trigger keywords are broad and overlap with common user requests such as ‘校对’, ‘检查文档’, and ‘proofreading’, which can cause the skill to activate in situations beyond the intended scope. Over-broad activation is dangerous because it may intercept unrelated tasks, increase unintended file-reading behavior, and cause the agent to apply this skill when the user did not intend document analysis.

Vague Triggers

Low

Confidence: 88% confidence
Finding: The skill description states what it does but does not clearly define activation boundaries or when it should not be used. This is risky because ambiguous routing can lead to accidental invocation for general editing, translation, or non-Chinese document tasks, reducing user control and increasing the chance of inappropriate tool use.

Static analysis

No suspicious patterns detected.