Aavegotchi GBM Skill

Security checks across malware telemetry and agentic risk

Overview

This is a disclosed Web3 auction helper that can sign real Base transactions, but its high-impact behavior is purpose-aligned and gated by dry-run and explicit broadcast instructions.

Install only if you are comfortable letting an agent prepare and, when explicitly authorized, broadcast Base mainnet transactions. Use a dedicated low-balance wallet, keep DRY_RUN=1 by default, verify contract addresses and auction details before any send, and avoid granting broad approvals unless you understand and can revoke them.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • 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)

Description-Behavior Mismatch

Medium
Confidence
95% confidence
Finding
The skill metadata and top-level description scope the capability to view/create/cancel/bid/claim auctions, but the documentation also includes executable buy-now flows. This creates a scope mismatch that can mislead operators, policy engines, or allowlists into permitting actions they did not intend to authorize, increasing the chance of unauthorized fund-spending transactions.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal