Back to skill

Security audit

learning-dashboard-plugin

Security checks for vulnerabilities and agentic risk

Overview

This skill is a narrowly scoped Chinese learning-progress helper that updates two disclosed Markdown records, with some risk of accidental activation from casual phrases.

Install this only if you want Chinese-language end-of-lesson phrases to update your learning dashboard files. Consider asking the agent to confirm the target files before updating when using short phrases like "存档" or "下课".

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
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (4)

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The skill metadata and description advertise activation on short, common phrases like '更新进度' and '下课', which are plausible in normal conversation and can cause unintended invocation. Because the skill performs file modifications, accidental triggering can lead to unauthorized or incorrect updates to learning records without an explicit confirmation boundary.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
97% confidence
Finding

The listed triggers include broad everyday expressions such as '今天到这', '下课', and '记录一下今天学了什么', which can occur in casual dialogue and do not reliably indicate consent to modify files. In this skill, ambiguous activation is more dangerous because invocation leads directly to persistent updates in project documents.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

The skill instructs the agent to update specific files and even defaults to '全部更新' when the user says generic phrases like '存档' or '下课', but it does not clearly warn that invoking the skill will automatically modify project files. This creates a consent and transparency problem: users may believe they are ending a session conversationally while the agent performs persistent writes.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
87% confidence
Finding

The manifest description is written only in Chinese, which indicates the skill may be targeted to a single language/locale without offering user choice or documenting that it is intentionally region-specific. Under the policy, language or locale constraints should be opt-in or clearly justified.

Content

No source excerpt is available for this finding.