ClawHub
Back to skill

Security audit

高考志愿填报参谋

Security checks across malware telemetry and agentic risk

Overview

This gaokao admissions counseling skill is mostly purpose-aligned, but it persistently stores sensitive student and family financial details without clear privacy controls or consent guidance.

Review this skill before installing if the user may share student, ranking, location, or family-income details. Only use it if you are comfortable with those details being written to memory, and prefer clearing or disabling stored profile/search-history data when the counseling task is finished.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • 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
Findings (4)

Context-Inappropriate Capability

Medium
Confidence
90% confidence
Finding
The skill explicitly instructs the agent to invoke a local Python CLI using an absolute filesystem path, which crosses from content guidance into host-level tool execution. Even if intended only for data lookup, this expands the agent’s operational scope and creates risk of unintended local command execution, environment disclosure, or misuse if the path/script is replaced or behaves unsafely.

Vague Triggers

Medium
Confidence
84% confidence
Finding
The trigger keywords are broad enough to activate on many ordinary education-related conversations, causing the skill to engage outside narrowly intended scenarios. Overbroad activation increases the chance of unsolicited behavior, unnecessary data collection, and user confusion, especially when the skill also promotes external services and stores profile data.

Vague Triggers

Medium
Confidence
86% confidence
Finding
The manifest trigger configuration repeats broad keywords without contextual constraints, making accidental invocation likely at the platform level. In this skill, that matters more because activation can lead to collection of student and family financial information and redirection toward paid consulting services.

Missing User Warnings

Medium
Confidence
94% confidence
Finding
The skill collects sensitive personal data about students, including province, exam type, ranking, interests, and family income, and writes it to memory files without a clear retention, consent, or privacy notice. This creates privacy and compliance risk because users may not understand that educational and financial profile data is being stored persistently.

VirusTotal

62/62 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.