Lead Gen Website Builder (SEO PUR)

• Outcome: Review
• Latest audit: May 28, 2026, 5:54 AM
• Version:
  1.0.0

Overview

This is a coherent website-building skill, but it should be reviewed because its lead form and WhatsApp workflow handle personal lead data without fully enforcing the privacy controls the skill claims to provide.

Install only with review and hardening before production use. Replace the default contact form with a secured backend submission flow, remove console logging of personal data, add a visible privacy notice and consent/lawful-basis handling near the form, disclose UTM/analytics/CRM and WhatsApp recipients, and define retention, deletion, access-control, and anonymization rules for lead records and conversation logs.

SkillSpector

By NVIDIA

Vulnerability Patterns

• MCP Least Privilege: Underdeclared Capability, Wildcard Permission, Missing Permission Declaration
• MCP Tool Poisoning: Hidden Instructions, Unicode Deception, Parameter Description Injection
• 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

Findings (2)

Lp3

Medium

• Category: MCP Least Privilege
• Confidence: 89% confidence
• Finding: The skill instructs the agent to read bundled reference files and write multiple project artifacts, but it does not declare corresponding permissions. Hidden or undeclared file access weakens the security model because users and enforcement layers cannot accurately understand or constrain what the skill is allowed to do. In this context the access appears aligned with the stated website-building workflow, so the issue is transparency and governance rather than overtly malicious behavior.

Description-Behavior Mismatch

Medium

• Confidence: 89% confidence
• Finding: The form collects personal data (name, phone, email, message, and UTM attribution data) but on submission it only writes that data to the browser console and shows a success toast, creating a misleading impression that a compliant lead-handling workflow occurred. In a lead-generation and RGPD-compliance context, this can cause silent loss of user requests, accidental exposure of PII in local logs, and non-compliant processing because consent, notice, retention, and actual secure transmission are absent.

VirusTotal

51/51 vendors flagged this skill as clean.