ClawHub

Mermaid Architect

Security checks across malware telemetry and agentic risk

Overview

The skill appears to be a normal Mermaid diagram-authoring helper with no evidence of hidden execution, credential use, persistence, or data exfiltration.

Installers should expect this skill to help create and validate Mermaid diagrams. Review whether its trigger wording is acceptable for your agent, since it may activate on general requests like visualizing a process, but the supplied evidence does not show sensitive access or hidden behavior.

SkillSpector

By NVIDIA
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 (2)

Vague Triggers

Medium
Confidence
94% confidence
Finding
The trigger phrases are very broad and map to common conversational requests such as 'Draw this' and 'Visualize', which can cause the skill to activate in contexts where the user did not specifically ask for Mermaid generation. Overbroad activation increases the chance of unintended tool routing, prompt interference, or the skill taking over benign requests that should be handled by a general assistant instead.

Vague Triggers

Medium
Confidence
91% confidence
Finding
The manifest description includes activation cues like 'diagram', 'flowchart', 'sequence diagram', and 'visualize this process', which are broad enough to overlap with ordinary user requests. Because manifest metadata is often used for routing, this can cause accidental or excessive invocation of the skill, making misapplication more likely across unrelated tasks.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal