ClawHub
Back to skill

Security audit

edu-homework-grader

Security checks for vulnerabilities and agentic risk

Overview

This is a coherent local homework-grading skill with no hidden network, credential, persistence, or destructive behavior found, but it can handle sensitive student information.

Reasonable to install for local educational grading. Before using it with real students, minimize or anonymize names and IDs, confirm that any assistant/OCR/LLM provider used by your environment is approved for student data, keep access limited, set a deletion schedule for generated reports, and require teacher review for official grades.

SkillSpector

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
88% confidence
Finding
The README promotes processing homework submissions and generating per-student reports, which likely involves personally identifiable information and educational records, but it provides no privacy, retention, or consent guidance. In a K-12 context this is more sensitive than generic document processing because student data may be legally protected and misuse can lead to unauthorized exposure or inappropriate downstream sharing.

Missing User Warnings

Medium
Confidence
88% confidence
Finding
The Chinese README likewise encourages use on K-12 homework and report generation without any warning that submissions may contain sensitive student information. Because this skill is explicitly intended for minors and individualized score reporting, the missing privacy notice increases the risk of operators handling regulated or highly sensitive data without safeguards.

Static analysis

No suspicious patterns detected.