ClawHub
Back to skill

Security audit

Word文档工具V1（免费版）

Security checks for vulnerabilities and agentic risk

Overview

This skill is a straightforward Word document helper whose file-writing abilities match its stated purpose, with no evidence of hidden or deceptive behavior.

Install only if you want an agent to create or edit local Word documents. Use explicit output paths, check before overwriting existing files, and require confirmation before any delete or replace operation.

SkillSpector

Vulnerability Patterns
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Behavioral ASTexec() Call, eval() Call, Dynamic Import
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Findings (3)

Vague Triggers

Medium
Confidence
89% confidence
Finding
The skill uses very broad, everyday trigger phrases such as creating, editing, and formatting documents without clear scope boundaries, opt-in conditions, or exclusions. In an agent environment with read/exec/write capabilities, this can cause the skill to activate on ordinary user requests and perform unintended file operations, increasing the risk of surprising or unauthorized modifications.

Missing User Warnings

Medium
Confidence
91% confidence
Finding
The description advertises creation, export, modification, and deletion-style operations but does not warn that these actions can change or remove user files. This is dangerous because users may issue natural-language requests without realizing the skill can perform destructive local actions, especially in a tool-enabled agent runtime.

Missing User Warnings

Medium
Confidence
95% confidence
Finding
The skill explicitly exposes exec and write tools and includes sample code that saves a .docx file, yet it does not clearly disclose that local files will be written or changed. Because the skill can both execute code and write to disk, missing warnings and guardrails materially increase the chance of unintended filesystem changes or misuse.

Static analysis

No suspicious patterns detected.