Relay Protocol MCP Server
Enables cross-chain bridging, multi-chain token swaps, and real-time pricing across over 50 blockchain networks using the Relay Protocol REST API. It allows users to browse thousands of tokens, generate quotes, and monitor transaction status directly through natural language.
README
Relay Protocol MCP Server
Model Context Protocol (MCP) server for the Relay Protocol REST API enabling cross-chain bridging and token swapping operations.
Features
- Cross-Chain Bridging - Bridge tokens between 50+ blockchain networks
- Multi-Chain Swaps - Aggregate tokens from multiple chains into single destination
- Real-Time Pricing - Get current token prices across all supported chains
- Request Monitoring - Track execution status and transaction details
- Currency Discovery - Browse 1000+ supported tokens with filtering
- Zero Configuration - No API keys required (free public API)
- Type Safety - Full TypeScript support with comprehensive validation
Quick Start
Option 1: Local Setup (Claude Desktop)
# Install and build
pnpm install && pnpm build
# Run development server
pnpm dev
# Run production server
pnpm start
Cloud Deployment (SSE transport)
Deploy to Railway, Heroku, Render, or any cloud platform:
# Build the server
pnpm build
# Start SSE server (uses PORT from environment or 3000)
pnpm start:sse
# Or for development
pnpm dev:sse
See RAILWAY.md for detailed Railway deployment guide.
MCP Integration
Stdio Transport (Local - Claude Desktop)
Add to your Claude Desktop configuration file:
Config Location:
- macOS:
~/Library/Application Support/Claude/claude_desktop_config.json - Windows:
%APPDATA%\Claude\claude_desktop_config.json - Linux:
~/.config/Claude/claude_desktop_config.json
{
"mcpServers": {
"Relay Protocol": {
"command": "node",
"args": ["/absolute/path/to/relay-protocol-mcp-server/dist/index.js"]
}
}
}
Option 2: Cloud Deployment (Railway)
Deploy Steps:
- Push your code to GitHub
- Go to railway.app → "New Project" → "Deploy from GitHub repo"
- Select your repository (Railway auto-detects configuration)
- Get your deployment URL:
https://your-app.railway.app
Claude Desktop Configuration:
{
"mcpServers": {
"Relay Protocol": {
"transport": "sse",
"url": "https://your-app.railway.app/sse"
}
}
}
Test Your Deployment:
curl https://your-app.railway.app/health
# Response: {"status":"healthy","server":"Relay Protocol","version":"0.1.0","transport":"sse"}
Deployment Info:
- Free tier: $5/month credits (plenty for personal use)
- Memory usage: ~50-100MB RAM
- Auto-sleeps after 30 minutes of inactivity
- Railway CLI:
npm i -g @railway/cli(optional)
Available Tools (9 total)
| Tool | Purpose | Example Use Case |
|---|---|---|
relay_get_chains |
List supported blockchain networks | "Show me all chains Relay supports" |
relay_get_token_price |
Get real-time token prices | "What's the price of USDC on Ethereum?" |
relay_get_currencies |
Discover tokens with advanced filtering | "Show verified tokens on Arbitrum" |
relay_get_quote |
Generate bridging/swap quotes | "Quote bridging 100 USDC from Ethereum to Polygon" |
relay_swap_multi_input |
Execute multi-chain swaps | "Swap ETH + USDC from multiple chains to USDC on Base" |
relay_get_execution_status |
Track request progress | "Check status of my bridge transaction" |
relay_get_requests |
Monitor request history | "Show my recent cross-chain transactions" |
relay_transactions_index |
Index transactions for tracking | "Register this transaction for monitoring" |
relay_transactions_single |
Index specific transfers/wraps | "Track this specific transfer operation" |
Example Prompts
# Bridge tokens
"Bridge 100 USDC from Ethereum to Polygon"
"What chains can I bridge USDC between?"
# Multi-chain swaps
"Swap all my USDC from Ethereum and Polygon to ETH on Arbitrum"
"Consolidate my tokens from multiple chains into USDC on Base"
# Price discovery
"Show current ETH prices across all chains"
"What's the cheapest way to get USDC on Optimism?"
# Currency discovery
"Find all stablecoins available on Polygon"
"Show me verified tokens with bridging support"
Development
pnpm typecheck # Type checking
pnpm dev # Development with auto-reload
pnpm build # Production build
pnpm start # Start production server
Project Structure
src/
├── client/ # HTTP client and error handling
├── tools/ # 9 MCP tool implementations
├── types/relay.ts # Complete TypeScript definitions
├── config.ts # API configuration
├── index.ts # Stdio transport (local)
└── server.ts # SSE transport (cloud)
Troubleshooting
"No available tools" in Claude:
- Verify absolute path in config is correct
- Test server:
echo '{"jsonrpc": "2.0", "id": 1, "method": "tools/list"}' | node dist/index.js - Restart Claude Desktop after config changes
- Ensure Node.js >= 20.0.0 and
pnpm buildcompleted
Connection issues:
- Run
pnpm buildbefore starting - Check server starts without errors:
pnpm dev - Verify JSON syntax in Claude config file
Requirements
- Node.js >= 20.0.0
- pnpm package manager
API Reference
Full access to Relay Protocol REST API with TypeScript types and validation. The Relay Protocol supports:
- 50+ Blockchain Networks including Ethereum, Polygon, Arbitrum, Optimism, Base, etc.
- 1000+ Tokens with real-time pricing and bridging support
- Free Public API with no rate limits or authentication required
- Production-Ready infrastructure handling millions in daily volume
License
MIT
推荐服务器
Baidu Map
百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。
Playwright MCP Server
一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。
Magic Component Platform (MCP)
一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。
Audiense Insights MCP Server
通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。
VeyraX
一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。
graphlit-mcp-server
模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。
Kagi MCP Server
一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
Exa MCP Server
模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。