Security audit

Video Channels Update Push

Security checks for vulnerabilities and agentic risk

Overview

This skill is a simple Markdown-based helper for checking public video-channel updates, with minor documentation and localization issues but no evidence of hidden or unsafe behavior.

Before installing, note that the instructions are Chinese-only and the channel-list filename should be clarified. The skill appears limited to maintaining a small list of public video channels and checking recent uploads; do not provide platform credentials to it unless you separately trust the environment and understand why login is needed.

A.I.G

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

SkillSpector

Vulnerability Patterns

MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
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

Findings (3)

Natural-Language Policy Violations

Medium

Confidence: 93% confidence
Finding: The manifest description and all user-facing instructions in SKILL.md are Chinese-only, which can amount to a language policy violation when no user opt-in or alternative language is provided. The file does not indicate that the skill is intentionally restricted to Chinese-speaking users or a China-specific compliance context.

Intent-Code Divergence

Low

Confidence: 94% confidence
Finding: The documentation says the channel list is stored in `video-channels.md` at L17, but later instructs users to read `youtube-channels.md` at L34 and even mentions editing `youtube-channels.md` at L22. This is an active contradiction in the skill's own instructions about where its managed channel data actually resides.

Natural-Language Policy Violations

Low

Confidence: 90% confidence
Finding: This markdown file presents all user-facing content in Chinese, including the title, description, table headers, and changelog. Under the policy, forcing a specific language without user opt-in can be a natural-language policy violation unless the locale constraint is explicitly justified, which is not stated here.

Static analysis

No suspicious patterns detected.