x402tools MCP Server
Provides AI agents with 10 pay-per-call utility tools (QR generation, DNS lookup, OCR, etc.) using USDC on Base via the x402 protocol, with agent's private key never leaving the agent.
README
x402tools MCP Server
A Model Context Protocol server that gives AI agents 10 pay-per-call utility tools. Pay with USDC on Base via x402 protocol — agent's private key never leaves the agent.
🛠 Tools
| Tool | Description | Price |
|---|---|---|
qr.generate |
Generate QR code from text/URL | $0.01 |
qr.generate_styled |
Artistic QR with shapes & gradients | $0.05 |
image.screenshot |
Website screenshot (dark mode, selectors) | $0.05 |
dns.lookup |
DNS records (A, AAAA, MX, NS, TXT, SOA) | $0.02 |
document.parse |
HTML/PDF → structured JSON | $0.01 |
security.screen |
Detect prompt injection / jailbreak | $0.03 |
email.validate |
Email validation + deliverability + risk | $0.03 |
document.render_pdf |
HTML/URL → PDF | $0.05 |
image.ocr |
Image → text via OCR | $0.05 |
prospect.enrich |
Name + domain → sales-ready buyer profile + cold-email opener | $0.50 |
🔌 Connect
Claude Desktop
Edit ~/Library/Application Support/Claude/claude_desktop_config.json (macOS) or %APPDATA%\Claude\claude_desktop_config.json (Windows):
Hosted (HTTP):
{
"mcpServers": {
"x402tools": {
"url": "https://mcp.x402tools.xyz/mcp"
}
}
}
Local (stdio with auto-pay convenience):
{
"mcpServers": {
"x402tools": {
"command": "npx",
"args": ["-y", "@x402-tools/mcp"],
"env": {
"WALLET_PRIVATE_KEY": "0x..."
}
}
}
}
Cursor
Edit ~/.cursor/mcp.json:
{
"mcpServers": {
"x402tools": {
"url": "https://mcp.x402tools.xyz/mcp"
}
}
}
Windsurf
Settings → MCP Servers → Add:
{
"x402tools": {
"serverUrl": "https://mcp.x402tools.xyz/mcp"
}
}
Cline (VS Code)
.vscode/mcp.json:
{
"mcpServers": {
"x402tools": {
"url": "https://mcp.x402tools.xyz/mcp"
}
}
}
Continue.dev
~/.continue/config.json:
{
"experimental": {
"modelContextProtocolServers": [
{
"transport": {
"type": "streamable-http",
"url": "https://mcp.x402tools.xyz/mcp"
}
}
]
}
}
OpenAI Agents SDK / Responses API
from openai import OpenAI
client = OpenAI()
response = client.responses.create(
model="gpt-4o",
tools=[{
"type": "mcp",
"server_label": "x402tools",
"server_url": "https://mcp.x402tools.xyz/mcp",
"require_approval": "never"
}],
input="Generate a QR code for https://x402tools.xyz"
)
Gemini / Google ADK
from google.genai import types
config = types.GenerateContentConfig(
tools=[types.Tool(mcp_server={"url": "https://mcp.x402tools.xyz/mcp"})]
)
💸 Payment Model
This server uses payment passthrough — the MCP server holds NO wallet. Each agent provides their own signed payment per call.
Flow:
- Agent calls a tool without payment → MCP returns
402 + payment requirements - Agent signs the payment locally with their wallet (hardware wallet, MPC, smart wallet, or private key)
- Agent retries the tool with
_paymentargument orPAYMENT-SIGNATUREHTTP header - MCP forwards the signed payment to the underlying x402 service
Easiest signing: use @x402/fetch to wrap your fetch — it handles the 402 → sign → retry loop automatically:
import { wrapFetchWithPaymentFromConfig } from "@x402/fetch";
import { ExactEvmScheme } from "@x402/evm";
import { privateKeyToAccount } from "viem/accounts";
const account = privateKeyToAccount(process.env.WALLET_PRIVATE_KEY);
const paidFetch = wrapFetchWithPaymentFromConfig(fetch, {
schemes: [{ network: "eip155:*", client: new ExactEvmScheme(account) }]
});
// Now calls to mcp.x402tools.xyz auto-pay on 402
Manual signing (for hardware wallets, MPC, smart accounts):
- Read
requirements.accepts[0]from the 402 response - Use
@x402/evm'sExactEvmScheme.createPaymentPayload(2, requirements)with yourClientEvmSigner - Wrap as v2 PaymentPayload:
{ x402Version: 2, payload, extensions: {}, resource, accepted } - Base64-encode the JSON
- Send as
PAYMENT-SIGNATUREHTTP header (or_paymenttool arg)
🏷 Protocol
- Network: Base mainnet (
eip155:8453) - Token: USDC (
0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913) - Receiver:
0xcEbbB82a183Faa69b929eD1F417aAFc63fa3b5b6 - Header:
PAYMENT-SIGNATURE(x402 v2)
🔍 Discovery Endpoints
- MCP server:
https://mcp.x402tools.xyz/mcp - Discovery JSON:
https://mcp.x402tools.xyz/.well-known/mcp - Health:
https://mcp.x402tools.xyz/health - MCP Registry:
io.github.acrylicfiddle/x402tools - npm:
@x402-tools/mcp - GitHub: acrylicfiddle/x402tools-mcp
🚀 Self-host
git clone https://github.com/acrylicfiddle/x402tools-mcp
cd x402tools-mcp
npm install
npm run build
PORT=4080 npm start
# MCP at http://localhost:4080/mcp
📦 Publishing to MCP Registry
brew install mcp-publisher
mcp-publisher login github
mcp-publisher publish
📜 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。