ClawHub
Back to skill

Security audit

36kr Hot Cn

Security checks across malware telemetry and agentic risk

Overview

This skill is low-risk to run locally, but it appears to return static sample headlines rather than live 36kr news.

Install only if static/sample 36kr-style headlines are acceptable. Do not treat its output as current news unless a future version implements real fetching and clearly identifies its source and freshness.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (3)

Description-Behavior Mismatch

Medium
Confidence
95% confidence
Finding
The skill advertises live monitoring/fetching of 36kr hot news, but the implementation returns hardcoded mock entries instead of current content. This is dangerous because downstream users or agents may make decisions based on fabricated or stale information while believing it is real-time external data, creating a trust and integrity failure.

Intent-Code Divergence

Medium
Confidence
94% confidence
Finding
The module docstring and function comments describe a real 36kr hot-news fetcher, but the code only serves embedded mock data. This mismatch increases the likelihood that users, orchestrators, or other agents will trust the output as authentic external intelligence, which can mislead workflows even without code execution or direct system compromise.

Vague Triggers

Medium
Confidence
92% confidence
Finding
The skill declares broad trigger terms such as '科技新闻' and '创业', which are common phrases users may say in many unrelated contexts. This can cause the skill to activate unintentionally, intercepting requests meant for other skills or general assistant behavior, which is a prompt-routing and scope-confusion risk even though the skill itself is only a news reader.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal