Lawb Chess

• Outcome: Review
• Latest audit: May 28, 2026 at 10:24 AM UTC
• Version:
  2.0.0

Overview

This skill is openly for on-chain wagered chess, but it gives an agent wallet-transaction and live public-state authority without clear per-action user confirmation.

Review before installing. Use a dedicated low-balance wallet, verify contract and token addresses independently, approve only the exact wager amount, and manually confirm every wallet transaction. Do not let an agent post chat, change profile data, update leaderboard fields, or join wagered games unless the chain, token, amount, opponent, and message content are explicit.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Privilege Escalation: Excessive Permissions, Sudo/Root Execution, Credential Access
• Supply Chain: Unpinned Dependencies, External Script Fetching, Obfuscated Code
• Excessive Agency: Unrestricted Tool Access, Autonomous Decision Making, Scope Creep

Findings (1)

Missing User Warnings

Medium

• Confidence: 91% confidence
• Finding: The skill instructs agents to connect wallets, approve tokens, place wagers, and write to a live Firebase game state without an upfront safety gate requiring explicit user confirmation of financial risk and live-state modification. In an agent context, this can lead to unintended on-chain transactions, token approvals, or game actions that spend funds or alter public state based on ambiguous prompts.

VirusTotal

66/66 vendors flagged this skill as clean.