ClawHub
Back to skill

Security audit

bili-tech-video-automation

Security checks for vulnerabilities and agentic risk

Overview

This documentation-only skill is not malware, but it guides design of an automated video acquisition and cross-platform publishing workflow that needs careful review for authorization, copyright, and platform-rule compliance.

Install only if you want PRD guidance for authorized video-processing and distribution products. Before using it to shape implementation work, require documented content rights, official platform API authorization, clear rate-limit compliance, human or policy gates before public posting, and removal of any watermark-removal or IP-pool behavior that would bypass platform or copyright rules.

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 (2)

Description-Behavior Mismatch

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

Although framed as a product-management/PRD workflow, the skill concretely specifies an end-to-end system for scheduled content acquisition, downloading, processing, subtitle generation, and automatic cross-platform publishing. That makes it operationally actionable for building automation that can violate platform Terms of Service, enable unauthorized redistribution, and reduce friction for large-scale content republishing despite the surrounding compliance language.

Content

No source excerpt is available for this finding.

Intent-Code Divergence

Medium
Category
Not specified by scanner
Confidence
96% confidence
Finding

The document claims it does not include executable crawling or auto-publishing behavior, but later provides detailed guidance for designing and operating those exact capabilities. This contradiction can mislead reviewers and lower scrutiny while still enabling development of risky automation for scraping and reposting workflows.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.