Back to skill

Security audit

Auto Course Player

Security checks for vulnerabilities and agentic risk

Overview

This skill automates a university course site to complete video playback, and its broad automatic clicking and course-completion behavior should be reviewed before use.

Install only if you are authorized to automate this course platform and it complies with school rules and academic integrity requirements. Review the Tampermonkey match scope and disable or narrow the generic popup-confirmation logic before using it on any logged-in university pages.

Vulnerability Patterns
  • Insecure Skill Coding PracticesFinds exploitable flaws such as hardcoded secrets or command injection
  • 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
Findings (1)

T09 · Insecure Skill Coding Practices

Warning
Location
scripts/course-autoplay.user.js:168
Finding

Overbroad Userscript Scope and Unsafe Automatic Confirmation Clicking

Content
View full analysis
btn.textContent.trim() === '确定' || btn.textContent.trim() === '确定 ') .find(btn => { // 确保按钮在可见的弹窗内,而不是课程列表中 const rect = btn.getBoundingClientRect(); const isCentered = rect.top > window.innerHeight * 0.2 && rect.top < window.innerHeight * 0.8; return isCentered; }); if (confirmBtn) { logInfo('检测到通用弹窗,自动关闭'); confirmBtn.click(); return true; } ``` The console version contains substantially the same fallback: ```javascript // 策略 3: 通用 - 查找屏幕中央的"确定"按钮(避免误点课程列表) const centeredConfirmBtn = allButtons .filter(btn => btn.textContent.trim() === '确定' || btn.textContent.trim() === '确定 ') .find(btn => { const rect = btn.getBoundingClientRect(); const isCentered = rect.top > window.innerHeight * 0.2 && rect.top < window.innerHeight * 0.8; return isCentered; }); if (centeredConfirmBtn) { centeredConfirmBtn.click(); log('✅ 已关闭弹窗 (通用)'); return true; } ``` The bookmarklet applies even weaker validation because it accepts any matching page element whose text contains the confirmation label: ```javascript function handleDialog(){ const c ...[truncated 2864 chars]
Remediation
View remediation
Vulnerability Patterns
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Findings (10)

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

The guide explicitly instructs users to automate playback, completion detection, pop-up dismissal, and progression through online course content, but provides no warning that this may violate the learning platform’s terms, undermine course integrity, or trigger account enforcement. In context, this is not a memory-safety flaw but a genuine security/compliance abuse concern because it operationalizes bypass of normal participation expectations on a real educational platform.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
88% confidence
Finding

The manifest description and user-facing instructions are written as mandatory Chinese commands such as "告诉 Bill" and example invocations in Chinese, with no indication that other languages are supported. This creates a language/locale constraint without user opt-in, which matches the policy-violation category for natural-language requirements.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

The visible UI text, instructions, and control descriptions are entirely in Chinese, which effectively constrains use to a specific language. The file does not offer any user opt-in, alternate language, or documented justification that this is intended as a region-specific/localized-only skill.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

This script performs automated DOM interaction by clicking course items and confirmation buttons based on broad text and layout heuristics, without explicit user confirmation for each action or clear scoping guarantees. In a browser console context, that can cause unintended navigation, dismissal of important dialogs, or accidental interaction with unrelated page elements if selectors match unexpectedly or the page structure changes.

Content

No source excerpt is available for this finding.

Context-Inappropriate Capability

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The script intentionally detects and dismisses dialogs about learning status/time, which goes beyond passive autoplay and can bypass course prompts, acknowledgements, or attendance/interruption checks. In the context of an educational platform, this may automate user-presence controls and undermine platform policy or record-keeping, even though it does not directly exfiltrate data or execute arbitrary code.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
88% confidence
Finding

The dialog dismissal behavior operates automatically without prominent user consent or warning that course prompts may be skipped. This increases the risk that users unknowingly bypass important acknowledgements or compliance steps on the training platform, making the automation more dangerous in this specific context.

Content

No source excerpt is available for this finding.

Intent-Code Divergence

Medium
Category
Not specified by scanner
Confidence
97% confidence
Finding

The header comment says this is a 'core control script' used to analyze page structure, play videos, detect completion, and switch to the next lesson. However, the implemented code only parses and stores course progress in a markdown file and exposes simple command-line operations, with no page analysis or playback logic present.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
84% confidence
Finding

The README instructs users to invoke and control the skill using fixed Chinese phrases such as "开始自动播放网课", "暂停播放", and "继续播放". This is a natural-language locale constraint presented as the expected interaction model, but the document does not offer alternatives or state that Chinese is optional or required for a justified region-specific reason.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
76% confidence
Finding

The script name, description, notifications, and interface text are all hard-coded in Chinese, with no indication that the skill is intentionally limited to a Chinese-speaking audience or a specific regional deployment. Under the stated policy, forcing a specific language without opt-in can be a natural-language policy violation.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
97% confidence
Finding

The script formats timestamps with toLocaleString('zh-CN', { timeZone: 'Asia/Shanghai' }), toLocaleDateString('zh-CN'), and toLocaleTimeString('zh-CN', ...), which forces a specific language/locale and region setting in user-facing output. This is a natural-language policy concern because the file provides no opt-in, fallback, or documented justification for restricting output to Chinese locale formatting.

Content

No source excerpt is available for this finding.