ClawHub
Back to skill

Security audit

选题生成器

Security checks for vulnerabilities and agentic risk

Overview

This skill is a disclosed content-idea generator with optional search and saving features, and the bundled code only creates local output files.

Install only if you want an agent to generate content ideas using trend searches and optionally save them. Before using Tencent Docs, cron scheduling, video generation, or publishing-related companion skills, confirm the destination and review any private performance data, audience data, or competitor lists you provide.

SkillSpector

Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Findings (4)

Lp3

Medium
Category
MCP Least Privilege
Confidence
88% confidence
Finding
The skill advertises behavior that can save outputs to the local workspace and references helper scripts for persisting generated ideas, yet no explicit permission declaration is shown. Undeclared write capability weakens user consent and platform enforcement because a skill that appears read/generate-only can also create or overwrite local files.

Tp4

High
Category
MCP Tool Poisoning
Confidence
93% confidence
Finding
The top-level description frames the skill as an idea generator, but the body also includes exporting to local Markdown/Excel, saving to Tencent Docs, and chaining into downstream automation such as video generation. This mismatch can mislead users and reviewers about side effects, enabling broader data handling and action execution than expected.

Vague Triggers

Medium
Confidence
84% confidence
Finding
Trigger phrases like '给我一些选题建议' and '今天做什么内容' are broad, natural-language requests that could easily occur in ordinary conversation. Overbroad activation increases the chance the skill runs unintentionally, which is more concerning here because the skill can perform searches, save files, and invoke other skills.

Missing User Warnings

Medium
Confidence
90% confidence
Finding
The skill explicitly supports saving outputs to the local workspace and Tencent Docs, but it does not clearly warn users about what data may be stored, where it will go, or when writes/uploads occur. This creates privacy and consent risk, especially if user preferences, competitor lists, or historical performance data are included in generated artifacts.

Static analysis

No suspicious patterns detected.