ClawHub
Back to skill

Security audit

AI Mermaid Diagrams

Security checks across malware telemetry and agentic risk

Overview

This skill is a straightforward Mermaid diagram generator, but it renders diagrams through an external mermaid.ink service, so sensitive diagram contents should be treated carefully.

Install only if you are comfortable sending Mermaid diagram text to mermaid.ink for rendering. Avoid including secrets, real credentials, private IP maps, unreleased architecture, or sensitive authentication flows; use a local Mermaid renderer instead for confidential diagrams.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (3)

Description-Behavior Mismatch

Medium
Confidence
95% confidence
Finding
The skill sends user-provided diagram content to the external mermaid.ink service, but this data egress is not clearly disclosed in the manifest/description. That creates a real confidentiality and trust issue because architecture, network, auth, or data-flow diagrams often contain sensitive internal system details that would be transmitted off-box to a third party.

Intent-Code Divergence

Medium
Confidence
92% confidence
Finding
The documentation claims rendering is done locally with the `mmdc` CLI, but the instructions actually use `curl` to upload the Mermaid source to mermaid.ink. This discrepancy is dangerous because it can mislead users and downstream agents into believing sensitive content stays local when it is in fact transmitted externally.

Missing User Warnings

Medium
Confidence
94% confidence
Finding
The skill instructs saving files locally and making a network request to render the diagram without explicitly warning the user about these side effects. In this context, the diagrams may contain sensitive infrastructure, authentication, or API-flow information, so silent file creation and remote transmission materially increase privacy and security risk.

VirusTotal

63/63 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.