ClawHub
Back to skill

Security audit

Cross-Material Consistency Auditor

Security checks across malware telemetry and agentic risk

Overview

This skill coherently audits consistency across user-provided materials and produces reports without modifying the originals.

Installers should expect the skill to read the materials they provide and create audit outputs such as extracted claims, a diff matrix, an audit report, and unified-wording JSON. Use a dedicated output directory and avoid passing sensitive documents unless the audit agent is trusted for that content.

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 (2)

Lp3

Medium
Category
MCP Least Privilege
Confidence
91% confidence
Finding
The skill reads environment variables, reads arbitrary input materials, and writes multiple output artifacts, but no permissions are explicitly declared. That creates a trust-boundary problem: operators and users are not warned that the skill can access local files and env-backed data, increasing the chance of unintended data exposure or unsafe deployment assumptions.

Missing User Warnings

Low
Confidence
87% confidence
Finding
The skill explicitly generates multiple output files and copies a template into a report, but the markdown does not clearly warn the user that executing the skill will write to the filesystem. This is a real but low-severity safety issue because unexpected file creation can overwrite artifacts, leak derived content into shared workspaces, or violate read-only expectations in cautious environments.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.