ClawHub
Back to skill

Security audit

Mainline Anchor

Security checks across malware telemetry and agentic risk

Overview

This is a documentation-only workflow skill for keeping multi-agent work aligned, with one disclosed local status-file update expectation.

Install this if you want agents to follow a structured mainline-alignment workflow. Before using it in a shared repository, make sure the team is comfortable with agents creating or editing HEARTBEAT.md for status tracking, or ask agents to show the proposed entry before writing it.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Findings (1)

Missing User Warnings

Medium
Confidence
92% confidence
Finding
The skill explicitly requires updating HEARTBEAT.md on session exit, which authorizes a file write side effect without any user warning, confirmation boundary, or scope restriction. In an agent setting, this can normalize silent modification of workspace files and may lead to unintended state changes, audit noise, or corruption of shared coordination documents.

VirusTotal

63/63 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.