ClawHub
Back to skill

Security audit

hekouwang Harness Check

Security checks across malware telemetry and agentic risk

Overview

This skill is a disclosed Harness verification helper that runs project-provided checks and clearly separates automated results from manual review boundaries.

Install this only in projects that intentionally use the expected `.harness/` verification scripts. Before running working-tree checks, be aware they may inspect local Skill and Memory state; use CI mode when you want a fixture-based check that avoids those local dependencies. Do not approve `--fix`, commits, pushes, or external side-effect checks unless you intend those actions.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • 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)

Vague Triggers

Medium
Confidence
91% confidence
Finding
The skill’s activation description is very broad, covering common requests such as checking, explaining, diagnosing, maintaining, packaging, and open-sourcing a Harness workflow. That can cause the skill to trigger in situations beyond strict security verification, increasing the chance that an agent follows repository-provided scripts and governance flows in inappropriate contexts or without sufficient user intent validation.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.