Terms Of Service

• Outcome: Pass
• Latest audit: May 29, 2026 at 8:18 PM UTC
• Version:
  2.0.0

Overview

This skill is a basic Terms of Service drafting helper with local-only scripts and no evidence of exfiltration, destructive behavior, or credential use.

Install only as a boilerplate drafting aid. Review generated terms with qualified legal counsel before publishing, and avoid entering secrets or sensitive business details into the helper commands unless you are comfortable with that text being stored in the local terms-of-service data directory.

SkillSpector

By NVIDIA

Vulnerability Patterns

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

Findings (4)

Vague Triggers

Medium

• Confidence: 88% confidence
• Finding: The skill description and trigger language are broad enough that the agent could invoke this skill for generic requests about terms, agreements, or legal documents without clear boundaries on jurisdiction, purpose, or user intent. In a legal-document generation context, overbroad activation can cause the model to route users into automated legal drafting when they may need clarification, localized advice, or a different skill entirely, increasing the risk of misleading or inappropriate output.

Vague Triggers

Low

• Confidence: 80% confidence
• Finding: The audience statement says the tool is suitable for essentially anyone needing terms of service, which is overly broad and does not communicate limitations, qualifications, or appropriate deployment context. While not overtly malicious, this can encourage use by users with materially different legal needs and create unsafe overreliance on generic generated terms.

Missing User Warnings

Low

• Confidence: 84% confidence
• Finding: The add path writes arbitrary user-supplied content directly into a persistent local database file without any warning about retention, sensitivity, or storage location. In a skill context, users may provide secrets or personal data expecting transient processing, so silent persistence can cause unintended data exposure to other local users, backups, or later export operations.

Missing User Warnings

Medium

• Confidence: 93% confidence
• Finding: The _log function records command activity and arguments to a history file for every command, but this behavior is not disclosed in the help text. Because arguments may contain sensitive search terms, identifiers, or other private content, silent audit-style logging increases privacy risk and can expose data through local file access, backups, or support collection.

VirusTotal

66/66 vendors flagged this skill as clean.