Back to skill

Security audit

Find Reference Video via Frameset

Security checks for vulnerabilities and agentic risk

Overview

This skill is a straightforward reference-video search workflow with disclosed browsing and optional local download behavior.

Before installing, understand that this skill will browse video-reference sites and may download selected YouTube or Vimeo videos to ~/Downloads. Only download content you have rights to store, and monitor file sizes.

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
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (3)

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
90% confidence
Finding

The natural-language description and invocation instructions are written entirely in Chinese and specify a Chinese trigger phrase, implying a fixed language expectation. There is no indication that the user may choose another language or that the Chinese-only behavior is justified by a region-specific requirement.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The invocation trigger is very broad: any request like finding a reference clip for a topic can activate a workflow that performs web browsing, link extraction, and potentially downloading files. Overly permissive triggering increases the chance of unintended activation and can lead the agent to take networked and file-writing actions without sufficiently explicit user consent for each stage.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The skill instructs the agent to download videos directly to ~/Downloads/ using yt-dlp, but it does not present a prominent upfront warning that local files will be created. This creates a filesystem-impact risk: users may not realize the skill can write potentially large media files locally, which can cause unwanted persistence, storage consumption, or accidental handling of copyrighted/sensitive content.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.