🔍 VerdictSwarm MCP Server

The first crypto token scanner available via MCP. Give any AI agent the ability to analyze tokens for rug pulls, scams, and risk — powered by VerdictSwarm's 6-AI-agent consensus system.

Works with Claude Desktop, OpenClaw, Cursor, Codex, Windsurf, and any MCP-compatible client.

Why?

AI trading agents are making on-chain decisions with no risk analysis. VerdictSwarm MCP gives them instant access to:

• 6-agent consensus scoring — not one model's opinion, six independent AI agents debate the risk

• On-chain security audits — mint authority, freeze authority, honeypot detection, LP lock status

• Rug pull detection — holder concentration, bundle/sniper activity, contract age analysis

• Human-readable reports — markdown reports ready to share or embed

One tool call. Sub-second cached responses. No blockchain node required.

Quick Start

Install & Run

# Install from GitHub
pip install git+https://github.com/vswarm-ai/verdictswarm.git#subdirectory=mcp-server
VS_API_KEY=your_key verdictswarm-mcp

# Or with uvx (zero-install)
VS_API_KEY=your_key uvx git+https://github.com/vswarm-ai/verdictswarm.git#subdirectory=mcp-server

# Or clone and run
git clone https://github.com/vswarm-ai/verdictswarm.git
cd verdictswarm/mcp-server
uv run verdictswarm-mcp

Claude Desktop

Add to your claude_desktop_config.json:

{
  "mcpServers": {
    "verdictswarm": {
      "command": "uvx",
      "args": ["git+https://github.com/vswarm-ai/verdictswarm.git#subdirectory=mcp-server"],
      "env": {
        "VS_API_KEY": "your_key_here"
      }
    }
  }
}

Then ask Claude: "Check if this token is safe:

OpenClaw

mcpServers:
  verdictswarm:
    command: uvx
    args: ["verdictswarm-mcp"]
    env:
      VS_API_KEY: your_key_here

No API Key?

The server works without a key at free-tier limits (10 scans/day, basic scores only). Get a key at verdictswarm.ai for full access.

Tools

Example: Quick Score

User: What's the risk score for BONK?
Agent: [calls get_quick_score("DezXAZ8z7PnrnRJjz3wXBoRgixCa6xjnB7YaB1pPB263")]
→ Score: 74/100 (Grade B) — LOW-MEDIUM risk

Example: Rug Check

User: Is this new memecoin safe? 7xKXtg2CW87d97TXJSDpbD5jBkheTqA83TZRuJosgAsU
Agent: [calls check_rug_risk("7xKXtg...")]
→ DANGER
  🚨 Liquidity NOT burned or locked
  ⚠️ Mint authority active
  ⚠️ Token is less than 24 hours old
  ⚠️ Bundle/sniper activity detected

Resources & Prompts

Resources (reference data for agents):

• verdictswarm://help — Tool usage guide

• verdictswarm://scoring — Score interpretation (0-100 scale, grades A-F)

Prompts (pre-built workflows):

• should_i_buy(token_address) — Full investment analysis with recommendation

• portfolio_check(tokens) — Batch risk assessment across holdings

Supported Chains

Scoring Guide

Configuration

Architecture

MCP Client (Claude, Cursor, OpenClaw, Codex...)
    │
    │  MCP Protocol (stdio)
    ▼
┌──────────────────────────┐
│  VerdictSwarm MCP Server │  ← This package (thin wrapper)
│  FastMCP + Python        │
└──────────┬───────────────┘
           │  HTTP (httpx)
           ▼
┌──────────────────────────┐
│  VerdictSwarm API        │  ← Existing backend (Railway)
│  6 AI agents + on-chain  │
└──────────────────────────┘

The MCP server is a stateless wrapper — all intelligence lives in the VerdictSwarm API. This means:

• Lightweight deployment (no GPU, no blockchain node)

• Single source of truth for scan logic

• API-level rate limiting and caching already work

Development

git clone https://github.com/vswarm-ai/verdictswarm.git
cd verdictswarm/mcp-server
pip install -e ".[dev]"
pytest  # 47 tests, ~0.3s

License

MIT — see LICENSE.

Links

• Website: verdictswarm.ai

• API Docs: docs.verdictswarm.ai

• GitHub: vswarm-ai/verdictswarm

• MCP Spec: modelcontextprotocol.io

Built by