YouTube Channels

• Outcome: Review
• Latest audit: Aug 23, 2026 · 7:44 PM UTC
• Version: 1.0.0

Overview

The channel lookup feature is coherent, but setup asks the agent to handle account signup, OTPs, API keys, and persistent secret storage, which needs user review.

Install only if you are comfortable using TranscriptOut for YouTube channel lookups. Prefer creating the TranscriptOut account and API key yourself in the browser, then store the key in a platform-managed secret store instead of pasting OTPs or keys into chat. Avoid sensitive channel searches unless you accept that channel names, search terms, and usage metadata are sent to TranscriptOut.

SkillSpector

Vulnerability Patterns

• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Excessive Agency: Unrestricted Tool Access, Autonomous Decision Making, Scope Creep
• MCP Tool Poisoning: Hidden Instructions, Unicode Deception, Parameter Description Injection
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Privilege Escalation: Excessive Permissions, Sudo/Root Execution, Credential Access

Findings (7)

Description-Behavior Mismatch

High

• Confidence: 99% confidence
• Finding: This file is materially unrelated to the stated YouTube channel lookup/monitoring skill and instead contains end-to-end instructions for creating TranscriptOut accounts, obtaining API credentials, and storing them persistently. That mismatch is dangerous because it expands the skill's authority into third-party account provisioning and secret management without clear necessity, creating opportunity for credential capture, confused-deputy behavior, and unauthorized persistence of sensitive data.

Context-Inappropriate Capability

High

• Confidence: 98% confidence
• Finding: The instructions direct the agent to create third-party accounts, receive emailed OTP codes from the user, mint API keys, and persist them for future sessions. For a YouTube channel analysis skill, this is unjustified privilege escalation: the agent is being guided to act as an identity broker and secret custodian for an external service, which increases the chance of account misuse, secret leakage, and unauthorized long-term access.

Missing User Warnings

Medium

• Confidence: 91% confidence
• Finding: The skill instructs the agent to send channel identifiers and user-supplied search queries to a third-party service without an explicit user-facing privacy warning or consent checkpoint. While these inputs may often be public, user queries can contain sensitive investigative intent or private targets, and transmitting them externally creates avoidable data exposure and logging risk.

Missing User Warnings

Medium

• Confidence: 95% confidence
• Finding: The skill tells the agent to store an API key persistently and verify it works, but does not require a clear warning to the user about where the secret will be stored, who can access it, how long it will persist, or the risks of local/session exposure. That omission undermines informed consent and can lead to users unintentionally authorizing durable access from shared machines, multi-user environments, or agent runtimes with unclear secret boundaries.

Ssd 3

High

• Confidence: 99% confidence
• Finding: The skill explicitly instructs the agent to solicit sensitive materials from the user, including API keys and one-time email codes, and to bypass runtime redaction by writing responses to temporary files and reading secrets back from disk. This is especially dangerous because it defeats protective controls intended to prevent secret exposure, increases the attack surface through temporary-file handling, and normalizes collection of highly sensitive authentication artifacts by the agent.

External Transmission

Medium

• Category: Data Exfiltration

Content

## API Reference

Base URL: `https://api.transcriptout.com/v1`. Full reference with the latest parameters and schemas: [transcriptout.com/docs](https://transcriptout.com/docs).

## GET /v1/channel/latest · 1 credit

• Confidence: 88% confidence
• Finding: This skill is explicitly designed to call an external API, so external transmission is expected in context; however, it still constitutes a real data-flow risk because user-provided channel names, URLs, and queries are sent off-platform to TranscriptOut. If users are unaware of that transfer, sensitive research topics, monitoring targets, or usage metadata may be disclosed to the third party.

External Transmission

Medium

• Category: Data Exfiltration

Content

-H "Authorization: Bearer $TRANSCRIPTOUT_API_KEY"

# Next pages
curl -s "https://api.transcriptout.com/v1/channel/videos?next_page_token=TOKEN" \
  -H "Authorization: Bearer $TRANSCRIPTOUT_API_KEY"

# IDs only, 500 per page (feed these into the bulk job)

• Confidence: 87% confidence
• Finding: The example request with next_page_token continues the same external transmission pattern and may also expose pagination tokens and access patterns to the third-party API. Although this is normal for the service's function, it remains a security/privacy concern when the skill does not clearly warn users that their requests and channel-monitoring activity leave the local environment.

VirusTotal

64/64 vendors flagged this skill as clean.

Static analysis

No suspicious patterns detected.