ClawHub
Back to skill

Security audit

质量计划技能

Security checks for vulnerabilities and agentic risk

Overview

This is a coherent quality-plan document generator that locally creates the requested Markdown and HTML files, with no evidence of hidden exfiltration or destructive behavior.

Before installing, confirm you are comfortable with the skill storing quality-plan content locally, especially if the plan includes sensitive product, customer, supplier, or process details. Use clear filenames or a dedicated output folder to avoid overwriting existing documents.

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 (4)

Lp3

Medium
Category
MCP Least Privilege
Confidence
93% confidence
Finding
The skill explicitly instructs the agent to read reference files, write Markdown output, modify that file across review cycles, and later generate an HTML file, but it does not declare corresponding permissions or user-facing safeguards. Hidden file I/O increases risk because users may not realize the skill persists data locally or accesses repository files beyond the immediate prompt.

Vague Triggers

Medium
Confidence
72% confidence
Finding
The trigger conditions are broad ('用户需要制定产品质量计划、质量控制方案、质量管理文件'), which can cause over-activation in loosely related contexts and increase the chance the skill performs file operations when the user did not expect it. While not directly malicious, broad activation boundaries reduce user control and can lead to unintended processing of sensitive manufacturing or supplier information.

Missing User Warnings

Medium
Confidence
91% confidence
Finding
The skill directs the agent to write a Markdown file and later directly edit that file on user-requested revisions, but it does not require an explicit warning or consent before modifying the local filesystem. This is dangerous because it creates persistent side effects and could overwrite existing files or store sensitive product and customer data without clear user awareness.

Missing User Warnings

Low
Confidence
88% confidence
Finding
The skill instructs invoking a local Python script with user-derived parameters to produce HTML, but it does not require explicit disclosure or consent for command execution. Even though the command is local and appears intended for formatting, executing scripts expands the attack surface and can be abused if filenames or arguments are unsafe or if the script/template chain is modified.

Static analysis

No suspicious patterns detected.