Security audit

视觉文件整理免费版

Security checks for vulnerabilities and agentic risk

Overview

This file-organizing skill is not malicious, but it asks for broad local file and command authority while its triggers and safety statements are too loose for an automatic rename/move tool.

Install only if you are comfortable letting the agent inspect and reorganize local files. Run it first with dry-run on a small, explicit folder, check every proposed rename and destination, and avoid using it for unrelated conversion or content-extraction tasks.

SkillSpector

Vulnerability Patterns

Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
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
Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access

Findings (4)

Intent-Code Divergence

Medium

Confidence: 95% confidence
Finding: The skill states in its '安全红线' that it only has rename/move permissions, but the manifest declares broader capabilities including exec, read, and write. This mismatch can mislead users and calling agents about the real authority of the skill, increasing the chance of unintended file access or command execution beyond the documented safety boundary.

Vague Triggers

High

Confidence: 92% confidence
Finding: The top-level description says to use the skill for broad file processing, document conversion, format conversion, and content extraction, which extends well beyond the stated file-sorting purpose. Overbroad triggers can cause an agent to invoke this skill in inappropriate contexts, exposing user files to unnecessary reading, writing, or execution actions.

Vague Triggers

High

Confidence: 90% confidence
Finding: The dedicated trigger section uses vague wording like '需要文件处理、文档转换、格式互转、内容提取时使用,' which can activate the skill for many unrelated tasks. In an agent environment, ambiguous activation criteria raise the risk of unnecessary invocation and unsafe file operations outside the tool's legitimate scope.

Missing User Warnings

Medium

Confidence: 87% confidence
Finding: The skill performs rename and move operations that may be hard to undo, yet the description does not prominently warn users about irreversibility or the lack of rollback. Users may invoke it without understanding the risk of misclassification, broken references, or loss of original naming context.

Static analysis

No suspicious patterns detected.