Ethereum RPC MPC Server
An MCP server that bridges AI models with Ethereum blockchains via all JSON-RPC calls, enabling natural language queries for block numbers, balances, transactions, and smart contract data.
README
Ethereum RPC MPC Server
A TypeScript MCP server that leverages the MCP SDK to support all Ethereum JSON-RPC calls, enabling AI models to interact with blockchain data.
Overview
This project provides a Model Context Protocol (MCP) server that allows AI assistants like Cursor or Claude (or any MCP Client implementation) to interact with Ethereum blockchain data through standardized JSON-RPC interfaces. It acts as a bridge between AI models and EVM blockchains, enabling seamless access to on-chain data and functionality.
Installation
# Clone the repository
git clone git@github.com:Phillip-Kemper/ethereum-rpc-mpc.git
cd ethereum-rpc-mpc
# Install dependencies
yarn install
# Build the project
yarn build
# Start Inspector with default Ethereum RPC (you can change in the inspector settings on )
yarn inspector
Usage
Using with Cursor
To use this MPC server with Cursor:
- In Cursor, go to Settings > Cursor Settings > MCP
- Add a new MCP server with the following configuration:
- Name: Ethereum RPC
- Type: Command
- Command:
npx -y ethereum-rpc-mpc https://eth.llamarpc.com Ethereumor - Command:
node /path/to/ethereum-rpc-mpc/dist/server/index.js https://eth.llamarpc.com Ethereum
<img width="938" alt="Screenshot 2025-03-13 at 19 41 56" src="https://github.com/user-attachments/assets/f9e09d44-ea43-4ef2-9950-88300f6f9667" />
Analytics
You can also enable analytics using MCP Analytics Middleware to get an overview about number of requests, duration of requests as well as error reporting for all your tools and resources. To enable it, just pass the --analytics flag and additionally a path for the Sqlite File.
- Command:
npx -y ethereum-rpc-mpc https://eth.llamarpc.com Ethereum --analytics --db-path /Users/satoshi/Desktop/analytics.db
Starting the Server
You can start the server by providing an Ethereum RPC URL and an optional chain name:
yarn start [RPC_URL] [CHAIN_NAME]
# Using npx (without installing)
npx -y ethereum-rpc-mpc [RPC_URL] [CHAIN_NAME]
If no RPC URL is provided, it will default to "https://eth.llamarpc.com".
Example:
# Connect to Ethereum mainnet
yarn start https://eth.llamarpc.com Ethereum
# Connect to Zircuit
yarn start https://mainnet.zircuit.com Zircuit
Examples
Here are some examples of how to use the Ethereum RPC MPC server with Claude:
Getting the Current Block Number
What's the current block number?
Checking an Address Balance
What's the ETH balance of 0x742d35Cc6634C0532925a3b844Bc454e4438f44e?
Getting Contract Information
Is 0x6B175474E89094C44Da98b954EedeAC495271d0F a smart contract? If yes, what kind of contract is it?
Retrieving Transaction Details
Can you show me the details of transaction 0x5c504ed432cb51138bcf09aa5e8a410dd4a1e204ef84bfed1be16dfba1b22060?
Zircuit SLS (Sequencer Level Security) Methods
This server includes special support for Zircuit-specific RPC methods. These methods are not enabled by default but are automatically activated when connecting to a Zircuit endpoint (Chain ID: 48900).
The supported Zircuit methods include:
zirc_isQuarantined: Check if a specific transaction is quarantinedzirc_getQuarantined: Query all quarantined transactions with optional address filtering
These methods are particularly useful for monitoring transaction quarantine status.
Future Steps
Next to RPC, also support indexed API access
We plan to extend the functionality beyond basic RPC calls to include support for indexed blockchain data APIs. This will enable more efficient querying of historical data and complex on-chain analytics.
Multi Chain, Multi RPC Support
Future versions will support connecting to multiple chains and RPC endpoints simultaneously, allowing AI models to access data across different blockchains in a single session.
Client Implementation
A future enhancement will be to develop a client-side implementation that makes it easier to interact with this MCP server from various applications.
Server Analytics
We plan to add analytics capabilities to track usage patterns, popular RPC methods, and performance metrics. This will help optimize the server and provide insights into how AI models are interacting with blockchain data.
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。