ClawHub
Back to skill

Security audit

Vet Repo

Security checks across malware telemetry and agentic risk

Overview

Vet Repo is a local repository security scanner whose documented workflow is coherent and does not show hidden execution, persistence, destructive behavior, or active data exfiltration.

Reasonable to install if you want a local first-pass scanner for agent configuration risks. For private repositories, be aware that the documented command appears local-only, but the bundled shared module contains an unused npm/PyPI lookup helper; avoid separately invoking registry verification unless you are comfortable sending package names to those public registries.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • 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 (3)

Tp4

High
Category
MCP Tool Poisoning
Confidence
88% confidence
Finding
The skill is presented as scanning repository agent configuration files, but the finding indicates the underlying implementation also scans broader code and may perform live network access to external registries. That mismatch is dangerous because users may authorize the skill under a narrower trust model than its actual behavior, enabling unintended data exposure or outbound connections during a supposedly local config review.

Description-Behavior Mismatch

Medium
Confidence
93% confidence
Finding
This shared module includes live outbound registry lookups and post-processing logic that exceed the stated purpose of repository agent configuration pattern scanning. In a security scanner, unexpected network access increases attack surface, can leak scan metadata, and creates behavior users may not expect from an offline pattern database module.

Context-Inappropriate Capability

Medium
Confidence
91% confidence
Finding
The Verify_Package function performs outbound HTTP requests to PyPI and npm even though this file is presented as a shared pattern database for scanning agent configs. Unnecessary network access in a scanning component can expose repository/package names, fail unpredictably in restricted environments, and violate least-functionality expectations.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

Detected: suspicious.dangerous_exec, suspicious.dynamic_code_execution

Shell command execution detected (child_process).

Critical
Code
suspicious.dangerous_exec
Location
scripts/patterns.py:357

Dynamic code execution detected.

Critical
Code
suspicious.dynamic_code_execution
Location
scripts/patterns.py:350