ClawHub

Info Visualize

Security checks across malware telemetry and agentic risk

Overview

This is a local chart and report generator, but its browser-based HTML/SVG output can embed untrusted input without escaping.

Install only if you will use trusted or sanitized data for the HTML/SVG chart path. Prefer PNG output for untrusted text, and use the archive option only for reports you are comfortable retaining in a long-lived local folder.

SkillSpector

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

Context-Inappropriate Capability

Low
Confidence
79% confidence
Finding
The documentation recommends optionally archiving generated reports into a user memory directory unrelated to the core rendering task. This expands the skill's data-handling scope and can cause unintended persistence of possibly sensitive report contents into a cross-task storage location, increasing exposure and retention risk.

Context-Inappropriate Capability

Low
Confidence
88% confidence
Finding
The output path convention explicitly recommends writing to a fixed user archive path under a memory directory, which exceeds visualization needs and normalizes persistence into a sensitive location. If used broadly, this can leak generated content into long-lived stores without clear consent, making later unintended access or reuse more likely.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal