ClawHub
Back to skill

Security audit

Chinese Chess

Security checks across malware telemetry and agentic risk

Overview

This skill is meant to build Chinese chess games, but it tells the agent to read and reuse entire hard-coded local folders outside the skill package before asking the user.

Install only if you trust and control the referenced /Users/root009/projects/demos/g1/game2 and /Users/root009/projects/demos/g2/game2 directories. Before use, inspect those folders and prefer changing the skill to use bundled or user-provided template paths with an explicit confirmation step before reading or copying files.

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 (2)

Vague Triggers

Medium
Confidence
88% confidence
Finding
The trigger phrases for the 'improve' path are broad enough that a loosely related request about improving a chess/game experience could invoke this skill and cause it to read and apply code from a local directory. Because the workflow then instructs the agent to copy source directly, an accidental match can lead to unintended file access and code modification beyond the user's actual intent.

Missing User Warnings

Medium
Confidence
93% confidence
Finding
The skill tells the agent to read an entire local source tree and then apply that source directly to create or modify files, but only informs the user after the changes are made. This reduces transparency and can result in unexpected file writes or code transplantation from sensitive local paths without informed user approval.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.