AI Crypto Risk MCP Server

The Security Layer for Agentic Crypto Trading

Stop your AI agent from trading into scams. This MCP server provides real-time cryptocurrency risk analysis, protecting users and agents from honeypots, rugpulls, and malicious smart contracts.

Why Use This?

• Prevent Losses: Detect honeypots, blacklist functions, and rugpull patterns before trading
• Agent Safety: Give your AI trading agents a security checkpoint
• Liability Shield: Document due diligence for every transaction
• Multi-Chain: Ethereum, BSC, Polygon, Arbitrum, Base, Solana

Quick Start

npm install -g @aicryptorisk/mcp-server

Get your API key at: https://aicryptorisk.com/api-keys

Claude Desktop Configuration

Add to claude_desktop_config.json:

Mac: ~/Library/Application Support/Claude/claude_desktop_config.json Windows: %APPDATA%\Claude\claude_desktop_config.json

{
  "mcpServers": {
    "aicryptorisk": {
      "command": "npx",
      "args": ["@aicryptorisk/mcp-server"],
      "env": {
        "AICRYPTORISK_API_KEY": "acr_your_key_here"
      }
    }
  }
}

Restart Claude Desktop. Done.

Tools Available

analyze_crypto_risk

Instant risk analysis for any token.

Input:
  - contract_address: "0x..." (required)
  - chain: "ethereum" | "bsc" | "polygon" | "arbitrum" | "base" | "solana"
  - coin_name: "Token Name" (optional)

Output:
  - Risk score: 0-100 (higher = more dangerous)
  - Verdict: Low Risk / Elevated Risk / High Risk / Critical Risk
  - Red flags: Honeypot, blacklist, hidden owner, high tax, etc.
  - Positive signals: Verified source, renounced ownership, etc.

request_deep_research

Comprehensive 4-pillar analysis delivered via email:

1. On-Chain Security
2. Social Media Intelligence
3. Institutional Interest
4. Off-Chain Investigation

get_supported_chains

Returns list of supported blockchain networks.

Example Prompts

Once configured, just ask Claude:

"Check if this token is safe: 0xdAC17F958D2ee523a2206206994597C13D831ec7"

"Analyze the risk of trading PEPE on ethereum"

"Before I swap, verify this contract: 0x..."

Pricing

| Plan | Scans/Month | API Calls/Day | Price | |------|-------------|---------------|-------| | Free | 1/day | 10 | $0 | | Monthly | 75 | 50 | $9/mo | | Yearly | 150 | 100 | $69/yr | | Lifetime | 200 | 200 | $199 once |

Get Your API Key

The Safe-Trade Skill

For maximum protection, pair this MCP with our Safe-Trade Skill that enforces mandatory risk checks before any transaction:

# .claude/skills/safe-trade/SKILL.md
---
name: safe-trade
description: Check crypto risk before any transaction
---
BEFORE any crypto transaction, run analyze_crypto_risk.
If score > 75: REFUSE to proceed.

Security

• API keys are hashed with SHA-256
• Usage tracked per key for billing
• Rate limiting prevents abuse
• No sensitive data stored

Support

• Documentation: https://aicryptorisk.com/docs
• API Keys: https://aicryptorisk.com/api-keys
• Email: [email protected]
• GitHub: https://github.com/djangamane/montecrypto

License

MIT - Use freely in your agents and applications.

Built for the Agentic Economy. Protect your users. Protect your agents. Trade safe.