Security audit

Title Diagnostic Report - 综合诊断报告

Security checks for vulnerabilities and agentic risk

Overview

This skill only generates a Chinese academic-paper title diagnostic report and does not request access to files, tools, credentials, or persistent settings.

Before installing, expect this skill to respond in Chinese and to use a fixed diagnostic-report format for academic paper titles. It is low risk from the inspected artifact, but users who want English or lightweight title feedback may need to state that preference explicitly.

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 (2)

Vague Triggers

Medium

Confidence: 89% confidence
Finding: The trigger phrases are broad enough to match ordinary user requests about title checking, which can cause the skill to activate unexpectedly outside a narrowly scoped intent. This creates routing and prompt-selection risk: users may be pushed into a rigid diagnostic workflow when they asked for something simpler, potentially degrading reliability and causing unintended processing of user content.

Natural-Language Policy Violations

Medium

Confidence: 84% confidence
Finding: The skill metadata and examples prescribe Chinese-language usage/output without offering any language negotiation or justified locale restriction. In multilingual environments this can cause incorrect routing, unusable outputs, or user confusion, especially if the system activates the skill for users who did not request Chinese.

Static analysis

No suspicious patterns detected.