Security audit

课后记录助手

Security checks across malware telemetry and agentic risk

Overview

This skill is a disclosed lesson-recording workflow for teachers, with sensitive student-data handling and lesson-unit ledger updates that are mostly scoped and documented.

Install only if you are comfortable with a skill maintaining persistent student lesson records and course-unit balances in a shared workspace. Use aliases, verify consent settings before cross-skill sharing, review any lesson-unit deduction within the adjustment window, and manually check parent summaries before sending them through an appropriate channel.

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 (3)

Vague Triggers

Medium

Confidence: 94% confidence
Finding: The activation guidance uses very broad natural-language phrases such as '课后总结' and '学员学习轨迹' that are likely to appear in ordinary teaching conversations. In an agent ecosystem with shared workspace writes and downstream integrations, accidental activation can cause unintended collection, modification, or propagation of student data without the teacher deliberately invoking this skill.

Vague Triggers

Medium

Confidence: 91% confidence
Finding: The trigger table includes ambiguous examples like '记一下这节课', '看一下 [化名] 的学习轨迹', and '扣 [化名] 一课时', which overlap with adjacent skills handling dashboards, planning, and billing-like ledger actions. Because this skill can write lesson logs, update progress, and pre-deduct lesson units across connected skills, ambiguous routing increases the chance of unintended state changes and privacy-relevant data exposure.

Missing User Warnings

Medium

Confidence: 92% confidence
Finding: The parent-brief section explicitly encourages sharing a student's performance, emotional state, behavioral issues, and learning trajectory with parents, but it provides no privacy guardrails such as consent, minimum-necessary disclosure, age/guardianship checks, or warnings against transmitting sensitive observations through insecure channels. In an education context, these details can be sensitive personal data, and normalizing their disclosure without safeguards increases the risk of over-sharing, misdelivery, or non-compliant handling of student information.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.