ClawHub
Back to skill

Security audit

Webapp QA Loop

Security checks across malware telemetry and agentic risk

Overview

This skill is a disclosed browser QA workflow that records local evidence and only allows repairs, delivery, deployment, or rollback when the user explicitly authorizes them.

Install this when you want an agent to perform real-browser QA and keep a durable local evidence ledger. Before using repair or release mode, be explicit about the target environment and which actions are allowed; do not grant production, delivery, rollback, or credential/session access unless you intend those specific effects.

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
94% confidence
Finding
The skill instructs the agent to read repository files, run shell commands, write ledger artifacts, and potentially edit source or deploy, but it declares no explicit permissions. That mismatch weakens policy enforcement and informed consent because a host may treat the skill as low-privilege while the instructions clearly drive file, environment, and shell-capable behavior.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.