ClawHub
Back to skill

Security audit

Diagram生成器

Security checks for vulnerabilities and agentic risk

Overview

This skill is a disclosed diagram-generation helper, but users should treat its external connector and file-writing behavior with normal caution.

Install only if you intend to use the external diagram connector. Review the connector package and configuration, keep API keys out of project files, and direct outputs to known diagram paths; use new filenames unless you intentionally want to replace an existing diagram.

SkillSpector

Vulnerability Patterns
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • 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)

Intent-Code Divergence

Medium
Category
Not specified by scanner
Confidence
97% confidence
Finding

The skill documentation makes concrete security assurances such as protection against API key leakage, data leakage, unauthorized access, and code injection, but SKILL.md itself does not implement or enforce those controls. This can mislead users and downstream agents into overtrusting the skill, causing them to process sensitive data or execute the skill in riskier contexts without verifying the actual protections of the external connector service.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
88% confidence
Finding

The activation guidance is overly broad ('use when improving efficiency, automation, batch processing, workflow optimization'), which can cause the skill to be invoked for many unrelated tasks. In this skill, broad triggering is more dangerous because the declared tool set includes read, write, and exec, increasing the chance of unnecessary file access, directory creation, or execution in contexts where a diagram tool was not specifically intended.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The skill describes edit workflows that read existing files and regenerate outputs, and it also states that the server creates missing directories and can write to default or specified paths. Without a clear user-facing warning about overwrite behavior and filesystem side effects, users or orchestrators may unintentionally replace existing diagrams or allow writes in unexpected locations, especially when combined with broad activation and write/exec permissions.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.