ClawHub
Back to skill

Security audit

Mindmap Gen Tool Free

Security checks across malware telemetry and agentic risk

Overview

This mind-map skill does not show malware, but it asks for command and file-changing authority that is broader than its simple stated purpose.

Review before installing. Use it only in a workspace where reading documents, writing generated mind-map files, and optional Markmap-related commands are acceptable. Avoid granting broad command execution unless the skill is revised to specify exact allowed commands, output paths, and confirmation requirements.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Behavioral ASTexec() Call, eval() Call, Dynamic Import
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Findings (4)

Description-Behavior Mismatch

Medium
Confidence
95% confidence
Finding
The skill is presented as a mindmap generator, but the documentation broadens behavior to generic create/query/modify/delete operations via runtime parameters. That scope expansion creates ambiguity for the agent and can justify actions far beyond the advertised purpose, increasing the risk of prompt-driven misuse or unexpected file/system operations.

Context-Inappropriate Capability

Medium
Confidence
90% confidence
Finding
The skill declares exec capability even though its stated purpose is document/topic-to-mindmap generation, which does not inherently require arbitrary command execution. Exposing exec unnecessarily increases attack surface because natural-language prompts or future workflow extensions could invoke system commands with user-controlled input.

Vague Triggers

Medium
Confidence
86% confidence
Finding
The invocation examples rely on broad natural-language requests like generating maps from topics or documents without clearly limiting data sources, side effects, or allowed tool usage. In a skill with read/write/exec available, vague triggers can cause the agent to overreach and perform unintended actions while trying to satisfy the request.

Missing User Warnings

Medium
Confidence
92% confidence
Finding
The skill exposes write and exec-capable behavior in its metadata and docs without clearly warning users that it may alter files or execute commands. That omission is dangerous because users may invoke what appears to be a harmless visualization tool while unknowingly granting it system-impacting capabilities.

VirusTotal

VirusTotal findings are pending for this skill version.

View on VirusTotal

Static analysis

No suspicious patterns detected.