ClawHub

日语朗读作业批改模板

Security checks across malware telemetry and agentic risk

Overview

This is a Japanese reading-grading instruction skill with expected student-audio privacy considerations but no hidden code, installer, upload behavior, or unrelated authority.

Before installing this for real classes, confirm you are allowed to process student recordings and identifiers, minimize names or student IDs where possible, prefer local transcription, restrict access to generated transcripts and reports, and delete audio/transcript/report artifacts when they are no longer needed.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Findings (2)

Missing User Warnings

Medium
Confidence
92% confidence
Finding
The skill explicitly supports batch import of student audio together with names or student IDs and generates individual and class-level reports, but it does not warn users that this involves processing personal data and potentially sensitive educational records. In a school context, combining voiceprints, transcripts, scores, and identifiers increases privacy risk because users may upload regulated student data without informed consent or proper handling expectations.

Missing User Warnings

Medium
Confidence
90% confidence
Finding
The documented workflow converts uploaded audio into WAV files and transcribes it into text, creating derived artifacts that may contain personal or sensitive spoken content, yet the skill provides no user warning that this transformation occurs. This is dangerous because users may reasonably expect only scoring, not storage or generation of intermediate files and transcripts that expand the privacy and disclosure surface.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal