ClawHub
Back to skill

Security audit

pretext-layout

Security checks across malware telemetry and agentic risk

Overview

This skill is a coherent browser text-layout helper with a small disclosed demo scaffold and no evidence of hidden credential, network, persistence, or destructive behavior.

Install this if you want browser-focused @chenglou/pretext integration guidance. If you run the scaffold, point --out at a disposable or intended project directory because it creates or overwrites index.html and demo.mjs there; separately review the third-party npm dependency if supply-chain provenance matters to you.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • 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
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (1)

Lp3

Medium
Category
MCP Least Privilege
Confidence
87% confidence
Finding
The skill instructs the agent to run a scaffold script that writes files to a user-specified output directory, but the metadata declares no permissions. This creates a capability mismatch: an execution environment or reviewer may assume the skill is read-only when it can actually modify the filesystem, increasing the risk of unexpected file creation or overwriting during use.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.