Neon Nodes Upload

• Outcome: Review
• Latest audit: Aug 5, 2026 · 4:15 PM UTC
• Version: 1.0.0

Overview

This NFT minting skill is purpose-related but asks agents to handle raw wallet private keys and submit paid blockchain transactions with too little user control.

Review carefully before installing. This skill should not receive a raw private key; prefer a version that uses a wallet, hardware signer, or user-reviewed signed transaction flow, and that clearly discloses any package installation or command execution.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Excessive Agency: Unrestricted Tool Access, Autonomous Decision Making, Scope Creep
• MCP Tool Poisoning: Hidden Instructions, Unicode Deception, Parameter Description Injection
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Privilege Escalation: Excessive Permissions, Sudo/Root Execution, Credential Access

Findings (7)

Intent-Code Divergence

Medium

• Confidence: 97% confidence
• Finding: The skill claims it does not access the filesystem, yet its fallback path instructs installing packages into /tmp, which necessarily writes to disk. False statements about filesystem behavior undermine trust boundaries and can conceal a larger attack surface than the user expects.

Intent-Code Divergence

Medium

• Confidence: 97% confidence
• Finding: The skill claims it does not access the filesystem, yet its fallback path instructs installing packages into /tmp, which necessarily writes to disk. False statements about filesystem behavior undermine trust boundaries and can conceal a larger attack surface than the user expects.

Context-Inappropriate Capability

High

• Confidence: 99% confidence
• Finding: The skill directs the agent to install ethers at runtime with npm into /tmp if unavailable. Runtime dependency installation introduces a supply-chain risk and expands the skill's authority from simple API interaction into package retrieval and local code execution, which is not necessary for a narrowly scoped NFT minting skill.

Missing User Warnings

High

• Confidence: 99% confidence
• Finding: The skill instructs the agent to ask the user for a raw EVM private key without a strong user-facing warning about the extreme sensitivity of that secret. In context, this is especially dangerous because the key grants full control over funds and NFTs, and the skill is for minting an NFT where safer signing patterns are readily available.

Ssd 3

High

• Confidence: 100% confidence
• Finding: The skill explicitly solicits the user's private key and tells the agent to complete the mint flow using that secret. Requesting raw wallet credentials is fundamentally unsafe because possession of the key enables total theft of assets, transaction signing for arbitrary transfers, and reuse beyond the stated minting purpose.

Ssd 3

High

• Confidence: 99% confidence
• Finding: The skill repeatedly instructs the agent to ask for additional wallet private keys once a wallet reaches its mint cap. That pattern broadens the credential-harvesting behavior from a single secret to potentially many, making the skill materially more dangerous and strongly suggestive of abusive intent.

Autonomous Decision Making

Medium

• Category: Excessive Agency

Content

## Agent Behavior

- The user will provide their **EVM private key**. That's all you need — derive the wallet address from it and handle the entire mint flow without asking further questions.
- **Before starting**, open with exactly this format: *"Mint a Neon Node. Mint price is 0.00075 ETH + gas on Robinhood Chain. I need your EVM private key."* Do NOT round the mint price.
- **First mint is always single** — send `quantity: 1` (or omit `quantity`) on the first `/api/puzzle` call. Do NOT ask about batch upfront.
- **Step 1** (puzzle): derive wallet address from private key using `new ethers.Wallet(PK).address`, then `curl` POST to `/api/puzzle` with `{ wallet }` for single (or `{ wallet, quantity }` on subsequent batch mints).

• Confidence: 88% confidence
• Finding: without asking

VirusTotal

VirusTotal findings are pending for this skill version.

Static analysis

No suspicious patterns detected.