Xiaozhi Teach Lesson Log

Security checks across malware telemetry and agentic risk

Overview

The skill is a coherent lesson-log assistant, but it can update student records and lesson-hour balances from broad trigger phrases without a clear pre-write confirmation step.

Install only if you are comfortable with this skill managing student lesson records and lesson-hour balances. Configure it so record updates, lesson-hour deductions, and parent-facing briefs require explicit teacher confirmation, use aliases only, avoid speculative behavioral or emotional claims, and review all parent communication before sending.

SkillSpector

By NVIDIA

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

Confidence: 91% confidence
Finding: The trigger list includes broad, everyday phrases such as '学员今天学得怎么样', '这节课上得怎么样', and '下节课应该讲什么', which can easily appear in normal conversation and unintentionally activate the skill. In an agent environment, over-broad activation can cause unintended collection or structuring of student-related information, execution of lesson-log workflows, and accidental downstream sharing to connected skills or systems.

Vague Triggers

Medium

Confidence: 89% confidence
Finding: The trigger phrases include broad, natural expressions such as '课后总结一下', '记一下这节课', and '这节课上得怎么样', which can plausibly occur in ordinary conversation and cause unintended activation. In this skill, accidental activation is more dangerous than usual because the workflow can update learning records and pre-deduct lesson hours, potentially causing unauthorized state changes or incorrect student records.

Missing User Warnings

Medium

Confidence: 95% confidence
Finding: The skill advertises automatic write-back to other systems and a pre-deduction mechanism for lesson hours, but it does not present a prominent, centralized warning that user actions may change records across linked systems. Because the skill writes to dashboard, analyzer, and parent-communication flows, missing upfront disclosure and confirmation can lead to silent data integrity issues, billing/lesson balance errors, and downstream propagation of incorrect educational records.

Missing User Warnings

Medium

Confidence: 88% confidence
Finding: The template explicitly encourages reporting a student's emotional state, behavior problems, and possible causes to parents, but provides no guidance on data minimization, age-appropriate sensitivity, factual framing, or consent boundaries. In an education context, these observations are privacy-sensitive and subjective; careless disclosure can harm the student, create legal/compliance issues, or lead to inappropriate labeling and escalation.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal