Back to skill

Security audit

视频批量转录

Security checks across malware telemetry and agentic risk

Overview

The skill appears to download and transcribe videos as advertised, but it needs Review because it encourages using browser cookies for logged-in and paid sites without clearly warning that those cookies can act like account credentials.

Install only after reviewing the cookie workflow. Use it only for content you are authorized to access, prefer isolated or temporary browser profiles/accounts, keep cookie files out of repositories and shared folders, restrict their permissions, delete them after use, and review output directories because downloaded media and transcripts remain on disk.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (8)

Lp3

Medium
Category
MCP Least Privilege
Confidence
89% confidence
Finding
The skill documentation clearly instructs users to run Python scripts, read browser cookies, and write output files, yet the skill declares no permissions. This creates a transparency and least-privilege problem: users and the platform cannot accurately assess or gate file and shell access before execution.

Missing User Warnings

Medium
Confidence
82% confidence
Finding
The document explicitly promotes downloading from 1000+ sites and using browser-derived cookies for authenticated access, but it provides no warning about privacy exposure, account risk, or possible terms-of-service/legal restrictions. In the context of an automation skill, this omission can lead users to supply sensitive session cookies or use the tool against protected services without understanding the security and compliance implications.

Missing User Warnings

Medium
Confidence
93% confidence
Finding
The README instructs users to extract browser cookies and provide cookie files for authenticated sites, including streaming services, but does not warn that these cookies are highly sensitive session credentials. In an agent-skill context, this can normalize handling reusable authenticated tokens and increase the risk of account takeover, privacy loss, or accidental exfiltration if users store, share, or process cookie files insecurely.

Missing User Warnings

Medium
Confidence
91% confidence
Finding
The report explicitly promotes use of browser-derived cookies and cookie files for accessing login-gated or paid platforms, but it provides no warning about authorization boundaries, sensitive credential handling, or privacy risks. In a skill designed to process content from many platforms, this omission can normalize unsafe reuse of session tokens and encourage users to access content without clearly understanding consent, account, or data exposure implications.

Missing User Warnings

Medium
Confidence
95% confidence
Finding
The skill instructs users to extract cookies from local browsers or provide cookie files for logged-in services, but does not prominently warn that these are active session credentials. In this context, cookies can grant access to private accounts on platforms like YouTube, Netflix, or other authenticated services, making mishandling equivalent to credential exposure.

Missing User Warnings

Medium
Confidence
81% confidence
Finding
The skill advertises optional export to Feishu documents without warning that transcript content and notes may be sent to an external third-party service. Because transcripts may include proprietary, personal, or regulated content, undisclosed outbound sharing creates a meaningful privacy and data-handling risk.

Missing User Warnings

Medium
Confidence
95% confidence
Finding
The examples repeatedly instruct users to supply browser-derived cookies or cookie files for authenticated sites, but they do not warn that these artifacts are effectively bearer tokens that can grant account access if exposed. In a tool designed to process content from many sites, normalizing cookie export as routine workflow increases the chance users mishandle sensitive session data or store it insecurely.

Missing User Warnings

Medium
Confidence
97% confidence
Finding
The FAQ explicitly recommends exporting cookies for Netflix and paid content, which are highly sensitive session credentials tied to authenticated and potentially billing-linked accounts. Presenting this without warning or safer handling guidance can lead users to expose active sessions, creating risk of account takeover or unauthorized access if the cookie files are leaked.

VirusTotal

67/67 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.