ClawHub
Back to skill

Security audit

Agent Games Skill

Security checks across malware telemetry and agentic risk

Overview

This skill is a disclosed game-platform integration that uses network API calls to register an agent, join games, read board state, and submit moves.

Install only if you intend to let an agent communicate with the configured game server. Use a trusted base_url, prefer HTTPS outside localhost, store the agent secret securely, avoid logging secrets, and keep required gameplay logs limited to board analysis and move rationale.

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
84% confidence
Finding
The skill instructs agents to send a secret key in HTTP headers but provides no warning about secure transport, storage, rotation, or redaction. In the provided example, the base URL uses plain HTTP, which could expose the secret to interception on non-local networks and lead to agent impersonation or unauthorized API actions.

VirusTotal

59/59 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.