ClawHub

Canvas Design

Security checks across malware telemetry and agentic risk

Overview

This is an instruction-only visual design skill with no executable code, though users should approve any external font downloads.

Install if you want a skill that strongly guides an agent toward polished static artwork. Before allowing font downloads, approve the source, prefer bundled or well-known font providers, and review generated .md, .png, or .pdf files before publishing or sharing.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (3)

Context-Inappropriate Capability

Medium
Confidence
94% confidence
Finding
The skill explicitly instructs the agent to download whatever fonts are needed, which introduces unnecessary network access beyond the core purpose of creating static artwork. This expands the attack surface through untrusted external content, possible data egress, and nondeterministic behavior, especially since no source restrictions, validation rules, or user consent are defined.

Vague Triggers

Medium
Confidence
81% confidence
Finding
The invocation description is broad enough that the skill may trigger for many generic creative requests, increasing the chance it is selected when a narrower or safer skill would be more appropriate. In this file's context that primarily creates overreach and unintended execution of the skill's behaviors, including the font-download instruction, rather than directly enabling a severe exploit on its own.

Missing User Warnings

Medium
Confidence
96% confidence
Finding
The skill tells the agent to download fonts without warning the user that network activity may occur. This undermines transparency and consent, and may surprise users in restricted environments while also increasing exposure to untrusted remote resources.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal