ClawHub
Back to skill

Security audit

Document to Mindmap

Security checks across malware telemetry and agentic risk

Overview

The skill performs the advertised ProcessOn mind-map conversion, but it also adds runtime update, document upload, tracking, and cleanup behaviors that users should review before installing.

Install only if you trust ProcessOn and the referenced GitHub update path, and only use it with documents you are comfortable uploading to ProcessOn. Do not approve the force-update command unless you intend to modify the installed skill from that repository, and avoid cleanup options or temp/cache input files that contain anything you need to keep.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • 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 (4)

Description-Behavior Mismatch

High
Confidence
97% confidence
Finding
The version-check and update workflow is unrelated to the core task of summarizing documents into mind maps, yet it requires outbound network access and conditions normal operation on remote state. This broadens the skill's authority and creates an unexpected trust path where a content-processing tool can interrupt work and steer users toward code-update actions.

Context-Inappropriate Capability

Critical
Confidence
99% confidence
Finding
The skill instructs the agent to execute shell commands and perform self-updating from a GitHub repository using a forced install command. A document summarization skill should not have authority to fetch and install new code at runtime, because this creates a direct path to arbitrary code changes and supply-chain compromise.

Context-Inappropriate Capability

Medium
Confidence
96% confidence
Finding
The skill requires submitting generated markdown and likely extracted document contents to a cloud endpoint, which is broader than a local document-structuring tool would imply. This exposes potentially sensitive user documents, meeting notes, reports, or images to a remote service without prominent upfront disclosure in the main skill description.

Missing User Warnings

Medium
Confidence
94% confidence
Finding
The skill mandates network version checks and potential code-pull/update behavior without sufficient user-facing risk disclosure or meaningful consent. Users asking for document summarization would not reasonably expect routine remote checks and possible software modification, making this an unsafe expansion of scope.

VirusTotal

VirusTotal findings are pending for this skill version.

View on VirusTotal

Static analysis

No suspicious patterns detected.