热点追踪专家

• Outcome: Pass
• Latest audit: May 29, 2026 at 8:36 PM UTC
• Version:
  1.0.0

Overview

This skill is a disclosed web-research and article-writing workflow, with some finance-marketing and scheduling caveats users should understand before use.

Install this if you want a Chinese-language hot-topic research and article-generation workflow. Before using it, be aware that it sends tracked topics to web/search tools, may create Markdown or DOCX article files, and has a strong finance-marketing/investor-education workflow; review all financial content for compliance and do not treat it as personalized financial advice. Use scheduled runs only for non-sensitive topics you intentionally want monitored over time.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• 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
• Supply Chain: Unpinned Dependencies, External Script Fetching, Obfuscated Code

Findings (5)

Description-Behavior Mismatch

Medium

• Confidence: 95% confidence
• Finding: The manifest describes a generic hot-topic tracker, but the body of the skill materially narrows behavior into financial education, business lead-generation, and product-promotion workflows. This mismatch can mislead users and higher-level orchestrators into invoking the skill in contexts they did not intend, increasing the risk of unsuitable financial-marketing output and compliance issues.

Description-Behavior Mismatch

Low

• Confidence: 88% confidence
• Finding: The documentation introduces scheduled follow-up behavior and automatic continuation of prior topics that are not disclosed in the manifest. Hidden persistence/reuse behavior can surprise users, cause unintended processing of prior context, and create privacy and consent issues when the skill runs in automated settings.

Missing User Warnings

Low

• Confidence: 87% confidence
• Finding: The skill instructs the agent to save files and potentially convert them to DOCX without warning the user that workspace files will be created or modified. Silent file creation is a consent and transparency issue that can lead to clutter, accidental overwrites, or unexpected persistence of generated content.

Missing User Warnings

Low

• Confidence: 85% confidence
• Finding: The skill mandates web searches and fetching on user-provided topics without warning that those topics will be sent to external services. If users supply sensitive names, entities, or internal topics, this can unintentionally disclose private information to third-party search/fetch providers.

Missing User Warnings

Medium

• Confidence: 94% confidence
• Finding: Automatic reuse of the previous tracking topic in scheduled runs means prior user context may be processed again without fresh consent or visibility. This creates a meaningful privacy risk, especially when prior topics involve sensitive interests, organizations, or ongoing monitoring targets.

VirusTotal

62/62 vendors flagged this skill as clean.