Overseas Engineering Bd

• Outcome: Review
• Latest audit: Jun 6, 2026 at 11:01 AM UTC
• Version:
  1.0.0

Overview

This is a disclosed business-development reference skill, but it enables scraping, contact enrichment, CRM imports, and WhatsApp/email outreach without enough consent, privacy, or user-control guardrails.

Review carefully before installing. Only use this skill for authorized business-development work, and do not allow automatic UTOS installation or any scraping, contact enrichment, CRM write, Google Contacts import, WhatsApp checking, or outreach execution without explicit approval, lawful basis, source review, rate limits, opt-out handling, and platform-policy compliance.

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
• MCP Tool Poisoning: Hidden Instructions, Unicode Deception, Parameter Description Injection
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Privilege Escalation: Excessive Permissions, Sudo/Root Execution, Credential Access

Findings (9)

Intent-Code Divergence

Medium

• Confidence: 95% confidence
• Finding: The skill claims it is only a reference repository with no execution framework, yet it instructs automatic detection, installation, and loading of another skill on activation. That creates an implicit execution and dependency-acquisition path without clear user consent, which can expand privileges, change runtime behavior, and introduce supply-chain risk through the referenced UTOS skill.

Vague Triggers

Medium

• Confidence: 82% confidence
• Finding: The trigger phrases include broad terms like overseas engineering, Africa BD, and overseas合作伙伴 that may match ordinary business queries unrelated to this skill's high-risk data collection and outreach functions. Overbroad activation can cause the agent to enter a workflow involving scraping, contact enrichment, and outbound messaging in situations where the user did not explicitly request those actions.

Missing User Warnings

Medium

• Confidence: 93% confidence
• Finding: The description advertises scraping, OCR, background investigation, contact completion, and outreach via email/WhatsApp, but it does not present a clear warning about privacy, consent, compliance, or external communications. In this context, omission is dangerous because the skill is explicitly designed for collecting personal/business contact data and initiating outbound contact, which can trigger privacy violations, spam, or regulatory breaches if users are not warned and gated.

Missing User Warnings

Medium

• Confidence: 90% confidence
• Finding: The file enumerates workflows for web scraping, PDF/OCR extraction, API integration, contact enrichment, WhatsApp number checking, and outreach execution, but provides no accompanying guardrails on consent, lawful basis, rate limiting, robots.txt compliance, personal-data minimization, or anti-spam controls. In this business-development context, the omission increases the risk that operators will collect or process personal/contact data and perform automated outreach in ways that violate privacy expectations, platform rules, or local regulations.

Missing User Warnings

Medium

• Confidence: 93% confidence
• Finding: The skill explicitly includes contact enrichment, background investigation, and CRM import of partner/contact data, but provides no constraints for lawful collection, consent, minimization, retention, or access control. In this business-development context, the omission is dangerous because it operationalizes bulk collection and enrichment of personal contact information for outreach, increasing the risk of privacy violations, unauthorized profiling, and misuse of scraped personal data.

Missing User Warnings

Medium

• Confidence: 95% confidence
• Finding: The file describes automated outreach sequencing, WhatsApp number pre-checking, persona-backed message generation, and SOP-based execution without any anti-spam, consent, opt-out, or platform-policy safeguards. In the skill's context, this is more dangerous because it is not a generic communications aid but a full lead-gen and outbound pipeline that can enable large-scale unsolicited messaging, deceptive personalization, and account/platform abuse.

Missing User Warnings

Medium

• Confidence: 94% confidence
• Finding: The skill explicitly supports contact enrichment via Hunter.io/Apollo.io and outreach preparation using email, phone, and WhatsApp, but it does not include consent checks, lawful-basis requirements, notice obligations, or jurisdictional privacy safeguards. In a lead-generation workflow targeting identifiable individuals, this omission creates a real risk of privacy violations, unsolicited contact, and misuse of personal data at scale.

Missing User Warnings

High

• Confidence: 97% confidence
• Finding: The workflow includes WhatsApp number verification, Google Contacts import, and outreach preparation, but provides no warning about privacy expectations, platform terms, anti-spam rules, or account enforcement risks. This is dangerous because it operationalizes bulk identification and messaging of individuals through a personal communications platform, increasing the chance of unauthorized processing, harassment complaints, and account bans.

Ssd 4

Medium

• Confidence: 95% confidence
• Finding: The guidance recommends using real-person photos instead of company logos, maintaining multiple salesperson accounts, and presenting a crafted 'technical helper' persona to warm targets. In context, this goes beyond ordinary marketing enablement and encourages identity shaping designed to lower skepticism and increase response rates, which can facilitate deceptive social engineering and evasive outreach practices.

VirusTotal

64/64 vendors flagged this skill as clean.