英语学习助手

• Outcome: Pass
• Latest audit: May 29, 2026, 8:28 PM
• Version:
  1.0.0

Overview

This is a Chinese-language English tutoring skill made of markdown and JSON files, with no evidence of hidden code, credential use, destructive actions, or external data transfer.

Install this if you want a Chinese-language English learning assistant. Avoid pasting sensitive personal text for correction or translation unless you are comfortable with local study-history retention, and treat voice-recording exercises as optional.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Trigger Abuse: Overly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Privilege Escalation: Excessive Permissions, Sudo/Root Execution, Credential Access
• Supply Chain: Unpinned Dependencies, External Script Fetching, Obfuscated Code

Findings (6)

Vague Triggers

Medium

• Confidence: 88% confidence
• Finding: The README includes many generic trigger phrases such as asking to translate, check a sentence, or recommend reading material, which overlap with ordinary conversation and can cause the skill to activate when the user did not explicitly intend to use it. In an agent platform, overly broad activation can misroute user requests, expose local memory-backed learning data in the wrong context, or cause the assistant to follow the skill workflow instead of the user's actual intent.

Vague Triggers

Medium

• Confidence: 88% confidence
• Finding: The README instructs users to invoke the skill with a very broad phrase equivalent to 'help me learn English' or any English-learning-related topic. This can cause the skill to activate on common conversational requests, increasing the chance of unintended routing and overcollection or processing of user content that was not meant for this skill.

Vague Triggers

Medium

• Confidence: 91% confidence
• Finding: The example trigger phrase is extremely generic ('help me make an English study plan' / similar), which overlaps with normal assistant usage. In practice, this can lead to accidental invocation when a user is simply asking for advice, making the skill boundary unclear and potentially interfering with expected assistant behavior.

Vague Triggers

Medium

• Confidence: 94% confidence
• Finding: The trigger description is broad enough to activate on many generic conversations about English, translation, reading, or speaking, which can cause unintended routing to this skill. Overbroad activation increases the chance of context hijacking, user confusion, and accidental exposure of user content to a skill that was not explicitly requested.

Natural-Language Policy Violations

Medium

• Confidence: 90% confidence
• Finding: The skill metadata and description are written to operate in Chinese for English-learning tasks without stating that the user must opt in to that language/locale. This can lead to unexpected language switching, misunderstanding of instructions, and incorrect handling of user requests when the user's preferred language or locale differs.

Missing User Warnings

Low

• Confidence: 82% confidence
• Finding: The skill explicitly tells the user to record their voice, but it provides no notice about whether the recording is stored, transmitted, or processed. Even in a benign language-learning context, prompting users to create audio containing biometric and personal data without a privacy disclosure can create unnecessary privacy risk and undermine informed consent.

VirusTotal

65/65 vendors flagged this skill as clean.