Back to skill

Security audit

linkdigest-youtube-summary

Security checks for vulnerabilities and agentic risk

Overview

This skill is a disclosed LinkDigest wrapper that sends user-provided video links to linkdigest.dev for transcript and summary generation.

Install only if you are comfortable sending submitted links or share text to linkdigest.dev and using a paid LinkDigest API key. Consider using --max-credits for cost control; the inspected script also references undefined emit/fail functions, so test with a low-cost or free request before relying on it.

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
Vulnerability Patterns
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • 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
Findings (3)

Undeclared Tool Scope

Medium
Category
MCP Least Privilege
Confidence
91% confidence
Finding

The skill declares capabilities that imply environment access, shell execution, and outbound network use, but it does not explicitly scope or constrain those permissions via a permissions or allowed-tools declaration. This increases the attack surface because a host agent may grant broader-than-necessary execution authority, making misuse of the API key or unintended command/network behavior harder to audit and restrict.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
89% confidence
Finding

The description states the skill is for Chinese summaries and can directly translate to Chinese, presenting Chinese as the default language behavior. Because the file does not indicate a user-selectable language policy or opt-in for this locale preference, it may violate language/locale policy requirements.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

技能清单说明该技能用于处理 YouTube 链接、Shorts 及相关视频总结/字幕提取,但代码中的命令行描述明确宣称可读取“小红书/抖音/TikTok/YouTube/X/公众号文章链接”。这表明实际实现面向的是一个更通用的多平台 LinkDigest 客户端,而非清单所述的 YouTube 专用能力。

Content

No source excerpt is available for this finding.