ClawHub
Back to skill

Security audit

文档导出工具免费版

Security checks for vulnerabilities and agentic risk

Overview

This is a local Markdown export skill whose file writes and command use match its stated conversion purpose, with some loose documentation that users should treat carefully.

Install only if you want an agent to run a local Markdown export CLI and write converted output files. Use explicit input and output paths, avoid pointing outputs at important existing files unless you intend to replace them, and treat the broad CRUD and network troubleshooting language as documentation noise rather than permission for unrelated actions.

SkillSpector

Vulnerability Patterns
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Behavioral ASTexec() Call, eval() Call, Dynamic Import
  • 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)

Intent-Code Divergence

Medium
Confidence
89% confidence
Finding
The skill claims the free edition is fully local and requires no cloud or service calls, yet the error-handling guidance instructs users to test connectivity with ping and troubleshoot network/firewall/proxy issues. This inconsistency can mislead an agent into performing unnecessary network activity, expanding behavior beyond the declared trust boundary and undermining user expectations about local-only execution.

Description-Behavior Mismatch

Medium
Confidence
91% confidence
Finding
The skill is presented as a Markdown export utility, but its capability descriptions introduce broad create/query/modify/delete operation modes unrelated to document conversion. That mismatch weakens scope constraints and could justify unintended actions by an agent, including modifying files or performing operations users did not anticipate from an export-only tool.

Vague Triggers

Medium
Confidence
84% confidence
Finding
The activation guidance is excessively broad, using vague scenario keywords that can cause the skill to trigger for loosely related file-processing or content-manipulation tasks. Overbroad invocation criteria increase the chance an agent selects this skill in inappropriate contexts, leading to unnecessary command execution or file writes.

Missing User Warnings

Medium
Confidence
93% confidence
Finding
The instructions direct command execution and output file creation/modification, but do not provide a clear warning that local files will be written or overwritten. In an agent context with exec and write capabilities, this omission can cause silent filesystem changes, data loss, or modification of unintended paths if the user has not explicitly consented.

Static analysis

No suspicious patterns detected.