Back to skill

Security audit

Nexus Curriculum Designer

Security checks for vulnerabilities and agentic risk

Overview

This is a low-risk curriculum-design prompt skill with some overbroad analysis and citation language users should treat cautiously.

Before installing, treat this as a prompt template for curriculum planning rather than a verified expert system. Ask the agent to use real sources when citations matter, and review generated standards alignment, assessments, and accessibility adaptations before using them in an educational setting.

Vulnerability Patterns
  • Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
  • Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
  • Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
  • Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
  • Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs
Vulnerability Patterns
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (2)

Description-Behavior Mismatch

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The workflow describes a generic analysis engine that produces findings, severity classifications, impact/effort scores, and cross-validation rather than a curriculum-design process. This mismatch can mislead operators about the skill’s real behavior, increasing the risk of scope confusion, inappropriate delegation, and unsafe trust in outputs presented as structured expert analysis.

Content

No source excerpt is available for this finding.

Intent-Code Divergence

Low
Category
Not specified by scanner
Confidence
84% confidence
Finding

The skill promises source citations and a minimum of two independent sources without describing any evidence-gathering or source-access mechanism. This can create false assurances of verification, causing users to overtrust unsupported educational recommendations or fabricated citations.

Content

No source excerpt is available for this finding.