Natural-Language Policy Violations
Medium
- Category
- Not specified by scanner
- Confidence
- 84% confidence
- Finding
L007 明确将 language 设为 zh-CN，而全文未说明这是可选偏好，也未向用户提供其他语言或地区变体的选择。按规则，强制单一语言/地区而无用户选择机制，可能构成自然语言层面的语言/locale 策略问题。
- Content
Security audit
中文防AI味写作助手
Security checks for vulnerabilities and agentic risk
Overview
This is a Chinese writing-style helper with broad activation phrases but no code execution, credential access, network use, persistence, or hidden behavior.
Install this if you want Chinese text rewritten in a more casual, less AI-like style. Be aware that broad triggers like rewrite, polish, and content creation may invoke it for general Chinese writing requests, so review outputs for tone, accuracy, and appropriateness before using them.
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
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 (4)
Vague Triggers
Medium
- Category
- Not specified by scanner
- Confidence
- 95% confidence
- Finding
The trigger conditions are very broad and match common, everyday requests such as '帮我写一篇文章/文案' or '让文字更自然'. This can cause the skill to activate unexpectedly in many unrelated writing contexts, increasing the chance that it overrides user intent, applies unwanted stylistic transformation, or interferes with other safer/more specific skills.
- Content
Natural-Language Policy Violations
Medium
- Category
- Not specified by scanner
- Confidence
- 88% confidence
- Finding
The manifest sets the skill language to "zh-CN", and the description also presents the tool as Chinese-only, but there is no indication that users can opt into another language or that the locale restriction is required for a region-specific purpose. This can be a natural-language policy concern when a skill imposes a language/locale without documenting user choice or justification.
- Content
Vague Triggers
Medium
- Category
- Not specified by scanner
- Confidence
- 95% confidence
- Finding
The trigger list contains very broad, common phrases such as '改写', '润色', '口语化', and '内容创作' that are likely to match ordinary user writing requests. This can cause the skill to activate unintentionally and influence unrelated tasks, which is a security and safety concern because users may not realize a paid or behavior-modifying skill has been invoked.
- Content
Static analysis
No suspicious patterns detected.