ClawHub
Back to skill

Security audit

Markdown 排版工具箱

Security checks for vulnerabilities and agentic risk

Overview

This is a local Markdown utility that reads user-named documents and writes derived conversion outputs, with no evidence of hidden network, credential, persistence, or destructive behavior.

Install only if you want a local Markdown formatter/converter. Use it on explicit document paths, review generated files before publishing, and avoid running conversion commands on sensitive documents unless you are comfortable with derived output files being created next to the originals.

SkillSpector

Vulnerability Patterns
  • 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
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Findings (4)

Lp3

Medium
Category
MCP Least Privilege
Confidence
81% confidence
Finding
The skill advertises file-processing behaviors and explicitly references scripts that read and transform user-supplied files, but it declares no permissions. That mismatch can bypass user and platform expectations about filesystem access, increasing the chance of unintended file reads or overwrites when the skill is invoked.

Tp4

High
Category
MCP Tool Poisoning
Confidence
96% confidence
Finding
The code broadly fits the stated Markdown tooling purpose, especially formatting, WeChat conversion, TOC generation, and link extraction. However, it also implements additional user-facing capabilities not mentioned in the description: Markdown table to CSV conversion, CSV to Markdown table conversion, and document statistics. Additionally, the description mentions Xiaohongshu conversion and heading hierarchy checking, but the code does not implement those features. Because the code exposes notable capabilities beyond the declared description, this should be flagged as a mismatch.

Vague Triggers

Medium
Confidence
78% confidence
Finding
An overly broad activation description can cause the skill to trigger on generic formatting requests without clear user intent to modify files. In a skill that can read and write documents, ambiguous invocation increases the risk of accidental file transformation, unwanted content changes, or use in contexts the user did not intend.

Vague Triggers

Medium
Confidence
76% confidence
Finding
Free-form invocation examples without scoped triggers encourage opportunistic matching on natural language requests, which can activate file-mutating behavior too easily. Because the skill performs transformations and conversions, vague triggers raise the likelihood of unintended execution against the wrong document or in the wrong format.

Static analysis

No suspicious patterns detected.