Lido MCP Server

Lido MCP Server

Enables AI agents to perform Lido staking operations (stake, unstake, wrap, unwrap, check balances/rewards) via natural language.

Category
访问服务器

README

Lido MCP Server

Model Context Protocol (MCP) server for Lido Finance operations

Built by AOX (@AOXexchange, aox.llc) for The Synthesis Ethereum Agent Hackathon 2026.


Overview

This MCP server gives AI agents native access to Lido staking operations. Point Claude (or any MCP-compatible client) at this server and stake ETH from a conversation — zero custom code.

Transport: stdio (MCP standard — not an HTTP server) Wallet: 0x7e7f825248Ae530610F34a5deB9Bc423f6d63373


Available Tools

Tool Description Network
lido_stake Stake ETH for stETH Ethereum Mainnet
lido_unstake Request unstake (withdrawal queue) Ethereum Mainnet
lido_wrap Wrap stETH to wstETH Base
lido_unwrap Unwrap wstETH to stETH Base
lido_balance Check stETH/wstETH balances + live APY Mainnet + Base
lido_rewards Get staking rewards data + projections Mainnet + Base

All write operations support dry_run: true for safe testing.


Installation

git clone https://github.com/GeObts/lido-mcp.git
cd lido-mcp
npm install

Environment Setup

cp .env.example .env
# Edit .env and set AOX_BANKER_PRIVATE_KEY (never commit real keys)

The server loads the private key from (first match wins):

  1. AOX_BANKER_PRIVATE_KEY environment variable (set by MCP client config)
  2. .env file in the project root
  3. ~/.openclaw/.env (legacy AOX agent path)

Usage with MCP Clients

Claude Desktop

Add to your Claude Desktop config (claude_desktop_config.json):

{
  "mcpServers": {
    "lido": {
      "command": "node",
      "args": ["/path/to/lido-mcp/index.js"],
      "env": {
        "AOX_BANKER_PRIVATE_KEY": "0xYourPrivateKeyHere"
      }
    }
  }
}

Restart Claude Desktop. The Lido tools appear automatically.

Claude Code

claude mcp add lido node /path/to/lido-mcp/index.js

Example Conversation

User: "Stake 0.01 ETH via Lido"

Claude: Calls lido_stake with dry_run: true first

Simulation successful:

  • Stake: 0.01 ETH → ~0.01 stETH
  • Gas: ~0.015 ETH
  • Current APY: 3.2% (live from Lido API)

Proceed with the real transaction?

User: "Yes"

Claude: Calls lido_stake with dry_run: false

Transaction confirmed:

  • Hash: 0x...
  • stETH received: 0.00998
  • Now earning staking rewards

Live Proof — Dry Run Results

Real simulations executed on mainnet:

lido_stake (dry_run)

{
  "success": true,
  "dry_run": true,
  "operation": "stake",
  "amount": "0.01",
  "contract": "0xae7ab96520DE3A18E5e111B5EaAb095312D7fE84",
  "network": "ethereum-mainnet",
  "estimated_stETH": "0.01",
  "current_apy_percent": 3.2,
  "gas_estimate": "~0.015 ETH"
}

lido_balance (live query)

{
  "success": true,
  "address": "0x7e7f825248Ae530610F34a5deB9Bc423f6d63373",
  "total_stETH_equivalent": "0.0",
  "current_apy_percent": 3.2,
  "apy_source": "Lido API (live)"
}

Security

This server handles real funds.

  • Private keys stored in .env (gitignored) — never committed
  • Rate limiting: 10 calls/minute per tool
  • Input validation: 0.001–10 ETH per stake transaction
  • Dynamic gas pricing with 20% buffer
  • dry_run mode on all write operations
  • Live APY from Lido API (cached 5 min, falls back gracefully)

AOX Integration

This MCP server powers the Banker Agent in the AOX autonomous pipeline:

  1. Marketplace Agent sells leads → receives USDC
  2. Banker Agent swaps 20% of USDC → ETH
  3. This MCP server stakes ETH → stETH via lido_stake
  4. Treasury earns 3-4% APY on idle funds

Fully autonomous. Zero human intervention.


Hackathon

Event: The Synthesis — Ethereum Agent Hackathon 2026 Team: AOX (Agent Opportunity Exchange) Prize Track: Lido MCP ($5,000) ERC-8004: Both Marketplace and Banker agents registered on-chain Repo: https://github.com/GeObts/lido-mcp


License

MIT — See LICENSE


Built by AOX · aox.llc · @AOXexchange

推荐服务器

Baidu Map

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。

官方
精选
JavaScript
Playwright MCP Server

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。

官方
精选
TypeScript
Magic Component Platform (MCP)

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。

官方
精选
本地
TypeScript
Audiense Insights MCP Server

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。

官方
精选
本地
TypeScript
VeyraX

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。

官方
精选
本地
graphlit-mcp-server

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。

官方
精选
TypeScript
Kagi MCP Server

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。

官方
精选
Python
e2b-mcp-server

e2b-mcp-server

使用 MCP 通过 e2b 运行代码。

官方
精选
Neon MCP Server

Neon MCP Server

用于与 Neon 管理 API 和数据库交互的 MCP 服务器

官方
精选
Exa MCP Server

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。

官方
精选