ClawHub
Back to skill

Security audit

gaokao-advisor中国高考志愿规划建议

Security checks for vulnerabilities and agentic risk

Overview

This is a Chinese gaokao college-advising skill that only provides researched recommendations and does not install code, persist data, or request unusual privileges.

Before installing, understand that this skill may ask for a student's province, subject track, score or rank, target majors, and target regions, and it will use web search to produce admissions advice. Treat its recommendations as advisory only and verify all admissions data against official exam authority and university sources.

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
Category
Not specified by scanner
Confidence
92% confidence
Finding

The trigger description is broad enough to activate on generic education-related queries such as school choice or major recommendations, which can cause the skill to run outside clearly scoped gaokao-planning contexts. Over-broad invocation increases the chance of irrelevant data collection, user confusion, and the model giving high-stakes admissions-style advice when the user did not explicitly ask for it.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
89% confidence
Finding

The skill metadata and behavior are effectively hardwired to Chinese-language operation without stating a justified locale restriction or offering fallback language handling. This can mis-handle multilingual users, lead to misunderstandings in a high-stakes advising workflow, and cause the assistant to provide recommendations in a language the user did not request or fully understand.

Content

No source excerpt is available for this finding.