ClawPay MCP — Non-custodial x402 Payment Layer for AI Agents
@up2itnow0822
About ClawPay MCP — Non-custodial x402 Payment Layer for AI Agents
Non-custodial x402 MCP payment protocol for AI agents. Agents sign locally, no custodial infra. Supports EVM (Base) + Solana, auto-handles HTTP 402 payment flows, on-chain spend limits. Published in Coinbase x402 ecosystem (PR #1381).
Basic information
Config
Add this server to your MCP-compatible client using the configuration below.
{
"mcpServers": {
"clawpay-mcp": {
"command": "npx",
"args": [
"-y",
"clawpay-mcp"
],
"env": {
"AGENT_PRIVATE_KEY": "<YOUR_PRIVATE_KEY>",
"AGENT_WALLET_ADDRESS": "<YOUR_WALLET_ADDRESS>",
"CHAIN_ID": "8453"
}
}
}
}Tools
No tools detected
We auto-extract tools from the README. The maintainer can list them under a ## Tools heading to populate this section.
Overview
What is AgentPay MCP?
AgentPay MCP is a Model Context Protocol server that wraps the Agent Wallet SDK, enabling MCP-compatible AI clients (Claude Desktop, Cursor, Windsurf) to make on-chain payments with built-in spend limit enforcement. It is non-custodial, runs on the Base network (Mainnet and Sepolia), and supports the x402 HTTP 402 payment standard. It is designed for developers building AI agents that need to pay for APIs, tokens, or services autonomously.
How to use AgentPay MCP?
Install globally via npm install -g agentpay-mcp, set environment variables (AGENT_PRIVATE_KEY, AGENT_WALLET_ADDRESS, optionally CHAIN_ID and RPC_URL), and add the server to your MCP client’s configuration (e.g., Claude Desktop’s claude_desktop_config.json). The server exposes tools such as deploy_wallet, send_payment, x402_pay, check_spend_limit, get_wallet_info, queue_approval, and get_transaction_history.
Key features of AgentPay MCP
- 🔐 Non-custodial — you hold your keys via an NFT-owned smart contract wallet.
- 💸 On-chain spend limits cap per-transaction and per-period spending.
- ⚡ x402-native automatic handling of HTTP 402 Payment Required responses.
- 🌐 Operates on Base network (fast, cheap, EVM-compatible).
- 🧩 Fully compatible with any MCP client (Claude Desktop, Cursor, Windsurf, etc.).
- 📋 Over-limit transactions queue for your explicit on-chain approval.
Use cases of AgentPay MCP
- AI agents autonomously paying for premium API access via x402.
- Automated recurring payments within strict per-period budget limits.
- Agent-to-agent micropayments for services, data, or compute.
- Safe deployment of agent wallets with spend controls to prevent runaway costs.
FAQ from AgentPay MCP
What does “non-custodial” mean for AgentPay MCP?
You own the wallet via an NFT. The agent only holds a limited hot key that can spend within on-chain limits you set; you retain full control of the wallet.
What happens if the agent private key is compromised?
On-chain spend limits cap any potential loss. The agent can only spend within the configured per-transaction and per-period limits, and over-limit transactions are queued for your approval.
Which networks does AgentPay MCP support?
It supports Base Mainnet (chain ID 8453) and Base Sepolia testnet (chain ID 84532), both EVM-compatible L2 chains.
How do I deploy a new agent wallet?
Use the deploy_wallet tool with token_id, factory_address, and nft_contract_address. After deployment, set AGENT_WALLET_ADDRESS in your environment.
What is x402 and how does it work in AgentPay MCP?
x402 is a standard where an API returns HTTP 402 Payment Required. AgentPay MCP’s x402_pay tool automatically fetches the URL, interprets the 402 response, and sends the required payment (within a max_payment_eth cap) using the agent wallet.
More Other MCP servers
Reactive Resume
amruthpillaiA one-of-a-kind resume builder that keeps your privacy in mind. Completely secure, customizable, portable, open-source and free forever. Try it out today!
Nginx UI
0xJackyYet another WebUI for Nginx

EverArt
modelcontextprotocolModel Context Protocol Servers
Awesome Mlops
visengerA curated list of references for MLOps
🚀 Model Context Protocol (MCP) Curriculum for Beginners
microsoftThis open-source curriculum introduces the fundamentals of Model Context Protocol (MCP) through real-world, cross-language examples in .NET, Java, TypeScript, JavaScript, Rust and Python. Designed for developers, it focuses on practical techniques for building modular, scalable,
Comments