🛡️ RugScan AI — Solana Token Risk Analyzer (MCP Server)

RugScan AI is a lightweight, effective risk assessment tool for Solana tokens. It runs as an MCP server and leverages the Helius API to scan SPL tokens for on-chain red flags like centralized control, holder concentration, and missing authority renouncements. It's built to give degens and developers instant clarity and trust scores.

✅ Built for the DeMCP MCP Hackathon
🔗 Uses only Solana's public APIs
🛠 Written in TypeScript

🚀 Features

Analyze any SPL token on Solana using its mint address

Detect risk factors such as:

• Mint authority not renounced
• Freeze authority still active
• Top holder dominance (> 50%)
• Very few unique holders

Generates a 0–100 Risk Score

Returns a human-readable risk summary (LLM-ready)

Can run as an MCP Server or standalone CLI

📽 Demo Video

Watch the demo

⚙️ How to Use

1. Clone the Repo

git clone https://github.com/Don-Vicks/rug-scan.git
cd rug-scan
npm install
npm run build

Make sure Node.js and TypeScript are installed globally.

2. Configure with Claude Desktop

Locate the index.js build file:

• After building, go to the build folder
• Copy the full file path to index.js

Edit Claude Desktop MCP config:

Windows

{
  "mcpServers": {
    "rugscan": {
      "command": "node",
      "args": [
        "C:\\FULL\\PATH\\TO\\rug-scan\\build\\index.js"
      ]
    }
  }
}

macOS

{
  "mcpServers": {
    "rugscan": {
      "command": "node",
      "args": [
        "/absolute/path/to/rug-scan/build/index.js"
      ]
    }
  }
}

Replace the file paths with your actual local path to index.js.

Coming Soon

• NFT collection scanning
• Wallet reputation scoring
• dApp contract safety checks
• Broader DeFi protocol support

License

MIT