Back to skill

Security audit

文件

Security checks for vulnerabilities and agentic risk

Overview

This file organizer is not clearly malicious, but it asks for broad file-changing and command authority while documenting duplicate deletion and external/API behavior without enough scope or safeguards.

Review this before installing. Only use it on backed-up, non-sensitive folders, and require the agent to show a plan and exact file list before moving or deleting anything. Do not provide an API key or allow network/API use unless the publisher explains why it is needed and what data is transmitted.

Vulnerability Patterns
  • Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
  • Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
  • Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
  • Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
  • Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs
Vulnerability Patterns
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • 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
Findings (5)

Missing User Warnings

High
Category
Not specified by scanner
Confidence
95% confidence
Finding

The skill states that duplicate files may be deleted, but it does not require explicit confirmation, preview, or safe rollback before performing a destructive action. This creates a meaningful risk of accidental data loss, especially in automated agent workflows where actions may be executed with limited human review.

Content

No source excerpt is available for this finding.

Context-Inappropriate Capability

High
Category
Not specified by scanner
Confidence
96% confidence
Finding

A file organization skill has no clearly justified need for an API key or external service dependency, yet the skill instructs users to configure one. In an agent environment, this can normalize unnecessary secret exposure and enable silent transmission of file metadata or content to third-party services.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The invocation guidance is vague and overly broad about when the skill should be used, which can cause an agent to select it for loosely related tasks. In combination with read/write/exec permissions, ambiguous routing increases the chance of unnecessary exposure of local files or misuse of privileged actions.

Content

No source excerpt is available for this finding.

Context-Inappropriate Capability

Medium
Category
Not specified by scanner
Confidence
84% confidence
Finding

The documentation advertises command execution as a general capability even though the stated purpose is file organization. Overbroad execution authority can let the skill perform actions unrelated to organizing files, increasing the risk of arbitrary command abuse or destructive filesystem operations if the agent follows the skill literally.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

Medium
Category
Not specified by scanner
Confidence
90% confidence
Finding

The skill is presented as a local file auto-organizer, but the documentation also claims external API integration and network-oriented behavior without explaining why network access is needed. This mismatch increases the attack surface and can mislead an agent into granting broader capabilities than necessary, including possible unintended data egress.

Content

No source excerpt is available for this finding.