百度热榜监控 | Baidu Hot Monitor

• Outcome: Pass
• Latest audit: May 29, 2026 at 8:21 PM UTC
• Version:
  1.0.0

Overview

This appears to be a purpose-aligned Baidu hot-topics monitoring skill that fetches public trend data and saves local history, but users should be aware it creates persistent local files.

Install only if you want a skill that fetches Baidu hot-topic data and keeps local history. Use it in a workspace where local SQLite databases, logs, and HTML reports are acceptable, and review or delete the data directory when you no longer need the retained monitoring history.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Trigger Abuse: Overly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
• 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

Findings (4)

Lp3

Medium

• Category: MCP Least Privilege
• Confidence: 88% confidence
• Finding: The skill advertises only a trending-topics monitor, but the documentation clearly indicates shell execution, network retrieval, and local file/database writes without any declared permissions or trust boundary explanation. This can mislead users and hosting platforms about the skill's actual capabilities, increasing the risk of unexpected persistence or external access when the skill is invoked.

Tp4

High

• Category: MCP Tool Poisoning
• Confidence: 93% confidence
• Finding: The documented behavior extends beyond simple retrieval of Baidu hot topics into persistent collection, historical querying, logging, and HTML report generation. This is a meaningful expansion of functionality that changes the privacy and security profile of the skill, because users may invoke it expecting ephemeral lookup while it actually creates durable local datasets and reports.

Vague Triggers

Medium

• Confidence: 81% confidence
• Finding: The trigger list includes very broad terms such as '百度', '热搜', and 'baidu', which are common in ordinary conversation and can cause unintended activation. In this skill's context, accidental invocation is more concerning because the documentation also describes network fetches and persistent local storage, so a casual mention could trigger actions with side effects.

Missing User Warnings

Medium

• Confidence: 95% confidence
• Finding: The documentation explicitly describes collecting hot-topic data, saving it into a SQLite database, and generating local HTML output, but it does not provide a clear warning that invoking these features creates persistent artifacts on disk. Lack of clear disclosure can lead to unintentional retention of research activity, monitored keywords, or usage history, which is especially problematic on shared systems or managed agent environments.

VirusTotal

66/66 vendors flagged this skill as clean.