Back to skill

Security audit

yt-dlp 视频下载 (eye)

Security checks for vulnerabilities and agentic risk

Overview

This skill is a straightforward yt-dlp download helper whose file writes and network use match its stated purpose.

Before installing, understand that using this skill will run yt-dlp against user-provided video links and create downloaded files in ~/Movies by default. Confirm the destination, avoid unintended playlist downloads, and do not provide site cookies or account credentials unless you understand the privacy implications.

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
  • 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
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Findings (2)

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

Hard-coding --sub-lang zh-Hans assumes a subtitle language preference without clear user opt-in. While not directly dangerous like code execution, it can cause unintended downloads, unexpected behavior, and privacy or UX issues if the agent fetches additional data the user did not request.

Content

No source excerpt is available for this finding.

Missing User Warnings

Low
Category
Not specified by scanner
Confidence
92% confidence
Finding

The skill defaults to writing downloaded content into the user's filesystem (~/Movies) but does not clearly warn users that triggering the skill will create files locally. In an agent context, silent file creation can surprise users, consume storage, and increase risk when large or repeated downloads occur without explicit confirmation.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.