ClawHub
Back to skill

Security audit

Youtube Apify Transcript

Security checks across malware telemetry and agentic risk

Overview

This skill does what it advertises: it fetches YouTube transcripts through Apify, with disclosed token use, network access, and local caching.

Install only if you are comfortable sending requested YouTube video URLs to Apify and storing transcript caches locally. Use a dedicated Apify token with appropriate quota limits, choose a cache directory suitable for your privacy needs, and be careful with broad transcript requests or --clear-cache.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (4)

Lp3

Medium
Category
MCP Least Privilege
Confidence
86% confidence
Finding
The skill clearly uses sensitive capabilities: it reads an API token from the environment, performs outbound network requests to APIFY, and documents local cache reads/writes. If the skill framework expects explicit permission declarations, omitting them weakens reviewability and can cause users or agents to invoke a networked, stateful skill without understanding its access scope.

Missing User Warnings

Medium
Confidence
91% confidence
Finding
The README advertises transcript fetching through APIFY proxies and local caching, but it does not clearly warn users that video URLs and transcript-related requests are sent to a third-party service and that fetched data may be stored locally. This creates a transparency and privacy risk: users may unknowingly transmit sensitive watch targets or retain transcript content on disk, which is especially relevant in enterprise or regulated environments.

Missing User Warnings

Low
Confidence
78% confidence
Finding
The documented `--clear-cache` action is destructive because it deletes locally stored transcript data, yet the skill text shows no warning, scope description, or confirmation prompt. In an agentic context, a user or upstream tool could trigger it unintentionally, causing avoidable data loss or workflow disruption.

Vague Triggers

Medium
Confidence
85% confidence
Finding
The trigger phrases are generic enough to match common user intents such as requesting a YouTube transcript, which can cause this skill to be invoked more often than users expect. In this skill's context, unintended invocation is more sensitive because it can send user-supplied video requests to a third-party scraping service using a privileged API token and a cloud-IP bypass workflow.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.