ClawHub
Back to skill

Security audit

novel-master

Security checks for vulnerabilities and agentic risk

Overview

This is a stateful novel-writing helper that stores project files locally for continuity, with no evidence of hidden network access, credential use, or destructive behavior.

Install this if you want a local, stateful writing workflow. Before using it on private or proprietary fiction, be aware that it creates and updates files under ~/.qclaw/workspace/novels; review or delete that directory when you no longer want the project retained.

SkillSpector

Vulnerability Patterns
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • 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)

Lp3

Medium
Category
MCP Least Privilege
Confidence
93% confidence
Finding
The skill instructs the agent to create, read, and update files under a local workspace, but it does not declare permissions or communicate those capabilities explicitly. Undeclared file access increases the risk of unexpected local data modification or access, especially if the skill is auto-invoked or used in broader contexts than the user intended.

Vague Triggers

Medium
Confidence
84% confidence
Finding
The trigger description is broad enough to match common writing-related requests such as '写小说' or '创作故事', which can cause the skill to activate in situations where the user only wanted lightweight help. Because the skill also performs file-oriented workflow steps, overbroad triggering can lead to unintended local side effects and workflow takeover.

Missing User Warnings

Medium
Confidence
95% confidence
Finding
The skill specifies creation of a structured local project directory and repeated updates to state and tracker files, but it does not clearly warn users that their prompts, outlines, and story data will be persisted locally. This is dangerous because users may unknowingly store sensitive or proprietary content on disk, and the agent may write files without informed consent.

Static analysis

No suspicious patterns detected.