ClawHub
Back to skill

Security audit

Security Sentinel

Security checks across malware telemetry and agentic risk

Overview

This skill behaves like a workspace security scanner, with some rough edges but no evidence of exfiltration, destructive behavior, or hidden persistence.

Install only if you are comfortable with a tool reading project files to look for secrets and with npm audit potentially making network requests. Keep its output out of public logs, and note that the bundled scan.js has some undocumented read-only policy checks and the dependency lockfile should be updated.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (2)

Tp4

High
Category
MCP Tool Poisoning
Confidence
91% confidence
Finding
The documented purpose is a generic security scanner, but the analysis indicates additional policy-enforcement behavior such as checking for specific files, forbidden paths, and package policies unrelated to the stated scope. Hidden or under-disclosed enforcement logic is dangerous because it can be used to manipulate workspace assessments, trigger false critical findings, or impose opaque policy decisions users did not consent to.

Missing User Warnings

Medium
Confidence
91% confidence
Finding
The skill recursively reads workspace files and inspects their contents for secrets without any consent prompt, scope restriction, or disclosure to the user. In an agent-skill context, this creates a privacy and data-exposure risk because the tool can access sensitive project material such as source code, configs, and credentials beyond what a user may expect from a routine scan.

VirusTotal

No VirusTotal findings

View on VirusTotal