Tutor

• Outcome: Pass
• Latest audit: May 28, 2026, 1:04 PM
• Version:
  1.0.0

Overview

This tutoring skill is instruction-only and purpose-aligned, but users should handle its local learner records carefully, especially for minors.

Install only if you are comfortable with local learner records being created under ~/tutor/. For minors, get parent or guardian consent, keep the folder private, avoid unnecessary identifiers or sensitive inferences, review reports before sharing, and delete records when they are no longer needed.

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 (6)

Vague Triggers

Medium

• Confidence: 88% confidence
• Finding: The skill description is extremely broad and lacks clear trigger boundaries, which can cause it to activate in contexts beyond intended tutoring use. Overbroad invocation increases the chance of handling sensitive requests or being used in inappropriate educational, medical, legal, or high-risk situations without adequate safeguards.

Missing User Warnings

High

• Confidence: 95% confidence
• Finding: The skill explicitly instructs persistent storage of learner profiles, session logs, progress records, and reports, but provides no user-facing notice, consent flow, retention policy, or data minimization rules. Because the subject matter includes children and parent oversight, this can result in collection of sensitive educational and potentially child-related personal data without informed consent or adequate privacy controls.

Missing User Warnings

Medium

• Confidence: 92% confidence
• Finding: The session flow directs the agent to read existing learner files and write updates on every session without warning the user that local files will be accessed and modified. This creates a transparency and privacy risk because users may disclose information assuming a transient conversation, while the skill silently persists and reuses it across sessions.

Missing User Warnings

Medium

• Confidence: 93% confidence
• Finding: The skill defines generation of parent reports for a minor that include detailed educational progress, struggles, engagement, and potentially sensitive notes, but it provides no guidance on consent, data minimization, access control, retention, or safe sharing. In a tutoring context involving children, this increases the risk of unauthorized disclosure of student information and over-collection or inappropriate propagation of sensitive learner data.

Vague Triggers

Medium

• Confidence: 85% confidence
• Finding: The non-urgent escalation criteria are subjective terms such as 'anxiety patterns,' 'unusual behavior changes,' and 'persistently unhappy,' which can lead to inconsistent or overly broad reporting of minors' conversations to parents. In a tutoring context, this creates privacy and trust risks because normal academic frustration or mood variation could be treated as grounds for disclosure without clear thresholds, review, or user transparency.

Natural-Language Policy Violations

Medium

• Confidence: 88% confidence
• Finding: The policy states that a parent has a right to session summaries for minors, but it does not describe consent flows, age handling, legal basis, or jurisdiction-specific limits. That can result in unauthorized or excessive disclosure of a minor's educational and emotional conversations, especially where privacy expectations or child-data rules require narrower sharing.

VirusTotal

66/66 vendors flagged this skill as clean.