teaching-plan-writer

• Outcome: Pass
• Latest audit: May 29, 2026 at 8:21 PM UTC
• Version:
  1.0.0

Overview

This is a lesson-plan writing skill that uses web search as part of its stated purpose and contains no executable code, persistence, credential access, or destructive behavior.

Install only if you are comfortable with the assistant using lesson topics and revision requests as web-search terms. Do not include student personal information, confidential school materials, or private classroom details in prompts, and review generated lesson plans for curriculum accuracy before use.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Trigger Abuse: Overly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Privilege Escalation: Excessive Permissions, Sudo/Root Execution, Credential Access
• Supply Chain: Unpinned Dependencies, External Script Fetching, Obfuscated Code

Findings (3)

Vague Triggers

High

• Confidence: 92% confidence
• Finding: The trigger logic is so broad that the skill may activate on generic user input and immediately invoke search-driven behavior. This increases the chance of unintended data disclosure, misrouting unrelated requests into this skill, and excessive external querying without clear user intent.

Missing User Warnings

Medium

• Confidence: 88% confidence
• Finding: The skill instructs the agent to broadly search the web and provide a downloadable Word link, but it does not disclose that user-provided teaching content may be transmitted to external services or stored in generated artifacts. This creates a privacy and data-handling risk, especially if teachers include student information, unpublished materials, or internal school content in prompts.

Missing User Warnings

Medium

• Confidence: 90% confidence
• Finding: The revision workflow explicitly re-runs external searches using teacher modification requests, again without warning that those instructions may be shared with outside services. Follow-up edits often contain more specific classroom, student, or institutional details, so repeated transmission can compound privacy exposure.

VirusTotal

66/66 vendors flagged this skill as clean.