Security audit

plain-explanation

Security checks for vulnerabilities and agentic risk

Overview

This skill only changes explanation style into simple child-friendly stories and does not request code execution, credentials, persistence, or private data access.

Install this if you want quick simplified explanations. Avoid relying on it for contexts where exact terminology or full technical precision matters, and ask for a normal or detailed explanation when needed.

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 (3)

Vague Triggers

Medium

Confidence: 95% confidence
Finding: The trigger list includes very common conversational phrases such as '再简单点', '还是不懂', and '打个比方', which can easily appear in ordinary dialogue unrelated to explicit skill activation. This creates a real prompt-routing vulnerability: the skill may be invoked unintentionally and rewrite responses into an oversimplified style, causing loss of precision or inappropriate behavior in contexts where accuracy matters.

Natural-Language Policy Violations

Medium

Confidence: 83% confidence
Finding: The file uses absolute natural-language instructions such as "永远不要堆术语，永远不要解释定义，永远直接讲故事" and states that every explanation must follow a single child-oriented style. This rigidly enforces one communication register rather than offering an opt-in style choice at response time, which can conflict with language/locale policy expectations for user-controlled communication style.

Natural-Language Policy Violations

Low

Confidence: 69% confidence
Finding: The description says it will automatically rewrite explanations so a 5-year-old can understand them, and the README presents this as the sole output mode. This is a natural-language style constraint without any opt-in choice among explanation levels or registers.

Static analysis

No suspicious patterns detected.