ClawHub
Back to skill

Security audit

WebToSkill

Security checks across malware telemetry and agentic risk

Overview

The skill does what it claims, but it renders untrusted web pages with Chromium sandboxing disabled by default, which needs user review before installation.

Review this before installing in a non-containerized or sensitive environment. Use it only for public URLs you are allowed to crawl, expect the generated ZIP to contain copied page snapshots, and prefer running with Chromium sandboxing enabled, for example by setting PLAYWRIGHT_NO_SANDBOX=0 where supported.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • 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 (2)

Lp3

Medium
Category
MCP Least Privilege
Confidence
90% confidence
Finding
The skill instructs the agent to perform network access, read bundled references, and write multiple artifacts, yet the skill declares no explicit permissions. That mismatch can mislead operators or policy enforcement about the skill's true capabilities, increasing the risk of unintended data access, external requests, and filesystem modification when the skill is executed.

Context-Inappropriate Capability

Medium
Confidence
94% confidence
Finding
The script launches headless Chromium with `--no-sandbox` and `--disable-setuid-sandbox` by default unless an environment variable explicitly disables that behavior. This materially weakens browser isolation while the tool renders untrusted remote web content, so a browser compromise would have a much easier path to host-level impact in the environment running the skill.

VirusTotal

VirusTotal findings are pending for this skill version.

View on VirusTotal

Static analysis

No suspicious patterns detected.