Mermaid Visualizer

• Outcome: Pass
• Latest audit: May 29, 2026, 8:17 PM
• Version:
  1.0.1

Overview

This appears to be a diagram-generation skill with some broad activation wording, but no evidence of hidden, destructive, credential-stealing, or unrelated behavior.

Install only if you want an assistant to generate diagram artifacts from your prompts. Use explicit requests such as “create a Mermaid diagram” or “generate an Obsidian Canvas file,” and check where it writes files before letting it run on an important vault or workspace.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Trigger Abuse: Overly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Privilege Escalation: Excessive Permissions, Sudo/Root Execution, Credential Access
• Supply Chain: Unpinned Dependencies, External Script Fetching, Obfuscated Code

Findings (6)

Vague Triggers

Medium

• Confidence: 84% confidence
• Finding: The trigger words documented for the skills include generic terms like 'diagram', 'flowchart', and 'mind map', which can overlap with ordinary user requests and cause unintended skill activation. In an agent environment, overly broad activation can route unrelated prompts into a file-generating skill, increasing the chance of unexpected workspace changes or prompt-scope confusion.

Vague Triggers

Medium

• Confidence: 91% confidence
• Finding: The Mermaid trigger list includes highly ambiguous words such as 'visualize' and 'flowchart', which are common in normal conversation and not specific to this skill. Because this skill transforms input into output artifacts, accidental activation could expose user content to unintended processing and lead to unrequested file creation or misleading autonomous behavior.

Vague Triggers

Medium

• Confidence: 88% confidence
• Finding: Canvas trigger words like 'mind map' and 'visual diagram' are too generic to establish reliable activation boundaries for a file-creating skill. In the context of Obsidian workspace tooling, this can cause the agent to invoke the Canvas creator during ordinary brainstorming or note-organizing requests, resulting in unintended artifact generation in the user's environment.

Missing User Warnings

Medium

• Confidence: 78% confidence
• Finding: The README states that the skills create files such as Mermaid, Excalidraw, and Canvas outputs, but it does not prominently warn that using the skills may create or modify files in the user's workspace. In an agent setting, insufficient disclosure about side effects can lead to unanticipated file writes, making accidental activation more consequential and reducing informed user consent.

Vague Triggers

Medium

• Confidence: 83% confidence
• Finding: The trigger phrases listed for the Mermaid skill are broad terms such as '可视化' and '流程图', which can overlap with ordinary user requests and cause the skill to activate in situations beyond the author's intended scope. In an agentic environment, over-broad activation increases the chance of unintended prompt injection surface, incorrect tool selection, or unexpected transformation of user content.

Vague Triggers

Medium

• Confidence: 89% confidence
• Finding: The skill description is broadly phrased around common user intents like 'visualize concepts' and 'make diagrams from text,' which can cause the skill to trigger on a wide range of ordinary requests. In agent systems that auto-select skills from natural-language descriptions, this can lead to unintended invocation, overreach into unrelated tasks, and increased exposure of user content to this skill when a more specific tool should have been chosen.

VirusTotal

64/64 vendors flagged this skill as clean.