Security audit

案例智能匹配分析

Security checks for vulnerabilities and agentic risk

Overview

This skill is a localized case-matching assistant that reads its bundled case library and creates a report, with no evidence of hidden code execution, network access, credential use, or destructive behavior.

Before installing, understand that this skill is mainly useful for Chinese-language users of the AI超级个体 case library. If your situation includes sensitive business or personal details, ask the agent to answer in chat only or confirm the filename and destination before it writes the Markdown report.

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

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 (5)

Natural-Language Policy Violations

Medium

Confidence: 89% confidence
Finding: The description and trigger terms are presented entirely in Chinese and the output filename/template are fixed in Chinese, indicating a language-specific constraint. The file does not state that Chinese is optional or provide a user-selectable language preference, which creates a natural-language policy concern.

Vague Triggers

Medium

Confidence: 93% confidence
Finding: The activation list includes generic phrases such as “分析案例”, “案例学习”, “找一个案例”, and “有没有类似的案例”, which could easily appear in ordinary discussion and unintentionally invoke the skill. The file does not provide exclusion conditions or negative examples to clarify when these phrases should not activate the skill.

Missing User Warnings

Medium

Confidence: 95% confidence
Finding: The skill instructs the agent to write a Markdown file by default without explicit user consent, confirmation of save location, or clarity about what data will be persisted. This can cause unintended storage of user-provided business details or personal context, creating privacy and data-handling risks, especially in environments where file writes persist or are shared.

Natural-Language Policy Violations

Medium

Confidence: 94% confidence
Finding: The file content is entirely in Chinese and defines user-facing case titles, scenarios, keywords, and takeaways without any indication of locale negotiation or fallback. In a skill intended to match users to cases, this can effectively force Chinese output or bias retrieval toward Chinese-speaking users, reducing accessibility and causing the agent to ignore a user's preferred language.

Natural-Language Policy Violations

Low

Confidence: 93% confidence
Finding: The natural-language default prompt is written only in English even though the skill's display name and description are in Chinese, which suggests a language constraint without user opt-in. This can violate a language/locale policy when the skill implicitly directs users to interact in a specific language.

Static analysis

No suspicious patterns detected.