Security audit

YouTube Video

Security checks across malware telemetry and agentic risk

Overview

This skill is a disclosed YouTube management wrapper, but it can upload, modify, rate, report, or delete videos when used with your YouTube credentials.

Install only if you trust the yutu CLI and want an agent to operate on your YouTube account. Keep client_secret.json and youtube.token.json private, avoid committing them, and require explicit confirmation before uploads, public/privacy changes, subscriber notifications, ratings, abuse reports, or deletions.

SkillSpector

By NVIDIA

Vulnerability Patterns

Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access

Findings (6)

Vague Triggers

Medium

Confidence: 91% confidence
Finding: The manifest description advertises broad natural-language triggers for destructive actions such as deleting videos, including phrases like 'delete my video' and 'delete videos', without any indication of confirmation or scope restriction. In an agent setting, overly broad trigger phrases increase the chance of accidental invocation of destructive operations from ambiguous user prompts or prompt-injection-influenced tool selection.

Missing User Warnings

Medium

Confidence: 87% confidence
Finding: The skill clearly exposes delete capability but does not warn that deletion is destructive and potentially irreversible. In a tool-using agent ecosystem, missing safety language makes it easier for an LLM or user to treat deletion as routine, increasing the risk of unintended content loss.

Missing User Warnings

Low

Confidence: 80% confidence
Finding: The skill states that OAuth credentials and a cached token are required but provides no privacy or handling warning around these sensitive artifacts. Although this file does not exfiltrate them, normalizing their use without guidance can lead to insecure storage, over-sharing, or accidental exposure in agent workflows.

Missing User Warnings

Medium

Confidence: 90% confidence
Finding: The setup guide instructs users to create OAuth credentials and cache an access token locally, but it does not warn that these files are sensitive secrets that can grant access to the user's YouTube account. In a skill intended for operational use, this omission increases the chance that users will store, share, or commit these files insecurely, leading to account compromise or unauthorized API use.

Missing User Warnings

Medium

Confidence: 90% confidence
Finding: This documentation exposes a destructive command that permanently deletes YouTube videos but provides no warning about irreversibility, ownership checks, or any confirmation/preview guidance. In an agent-skill context, terse instructions for destructive operations increase the chance of accidental or overly broad deletion, especially if an agent executes user intent directly from the documented example.

Missing User Warnings

Medium

Confidence: 93% confidence
Finding: The documentation instructs users to upload local video and thumbnail files to YouTube but does not clearly warn that file contents, embedded metadata, titles, descriptions, tags, and scheduling information will be transmitted to an external service. In an agent skill context, this omission increases the risk of accidental exfiltration of sensitive local media or metadata because users may treat the command as a routine local operation rather than a networked publish action.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.