ClawHub
Back to skill

Security audit

漫剧剪辑合成

Security checks for vulnerabilities and agentic risk

Overview

This is a plain Markdown editing-guidance skill with no code execution, credential access, network calls, or persistence.

This skill appears safe to install for comic-drama post-production planning. Users should be aware that generic trigger words like editing, subtitles, and export may activate it in broader video-editing contexts, so workflow routing should ideally check that the task is actually a manju/comic-drama editing step.

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
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • 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 (1)

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The trigger list includes very generic terms such as ‘剪辑’, ‘后期’, ‘字幕’, and ‘导出’, which are common in ordinary user requests and can cause this skill to activate outside its intended scope. In an automated workflow, broad activation increases the chance of unintended instruction injection into downstream processing or incorrect tool orchestration, especially since the file declares itself as automatically invoked by manju-workflow.

Content

No source excerpt is available for this finding.