x402-agent-mcp
MCP server that enables AI agents to search, describe, and fetch x402 endpoints on Base or Solana, with automatic USDC micropayments.
README
x402-agent-mcp
Universal x402 MCP for AI agents — discover and pay for any x402 endpoint on Base or Solana.
Agents discover services, pay per call in USDC, and consume data — all autonomously. No API keys, no subscriptions, no signup. Just a wallet.
What It Does
Agent: "I need news data"
→ x402_search("news") → finds 2s.io
→ x402_describe("2s.io") → gets endpoint schema + price
→ x402_fetch("https://2s.io/api/news/search?q=x402&limit=3") → pays $0.003 USDC → gets results
The agent never sees wallets, private keys, or x402 protocol details. Just search, discover, fetch.
Tools (8)
| Tool | Cost | Description |
|---|---|---|
x402_search |
Free | Search x402 endpoints by keyword, category, or chain |
x402_list_categories |
Free | List all endpoint categories with counts |
x402_describe |
Free | Get detailed info for a specific service (paths, prices, schema) |
x402_discover_url |
Free | Discover any x402 service by URL via well-known files + auto-add to directory |
x402_health |
Free | Check if a service is live and responding with 402 |
x402_discover_urls |
Free | Batch discover multiple x402 services in parallel |
x402_crawl_directory |
Free | Crawl x402scan.com to discover new x402 services and auto-add to directory |
x402_fetch |
Endpoint price | Fetch any x402 endpoint — handles 402 payment on Base or Solana |
Multi-Chain Support
| Chain | Env var | Payment |
|---|---|---|
| Solana | SOLANA_PRIVATE_KEY |
USDC via @x402/svm |
| Base | EVM_PRIVATE_KEY or BASE_PRIVATE_KEY |
USDC via @x402/evm |
Chain is auto-detected from the 402 response. Override with chain parameter.
Spending Limits & Payment Logging
| Env var | Default | Description |
|---|---|---|
MAX_PAYMENT_PER_CALL |
0.50 | Reject any single call above this amount (USDC) |
MAX_DAILY_SPEND |
10.00 | Reject after cumulative daily spend exceeded (USDC) |
PAYMENT_LOG_PATH |
./x402-payments.jsonl | Path to payment log file (gitignored) |
Payments are logged to x402-payments.jsonl with timestamp, URL, chain, amount, tx hash, and status. Daily spend resets at UTC midnight.
Quick Start
1. Install
git clone https://github.com/dchu3/x402-agent-mcp.git
cd x402-agent-mcp
npm install
npm run build
2. Configure
# .env
SOLANA_PRIVATE_KEY=your-base58-solana-key
EVM_PRIVATE_KEY=your-hex-base-key
SOLANA_RPC_URL=https://mainnet.helius-rpc.com/?api-key=your-key
BASE_RPC_URL=https://mainnet.base.org
MAX_PAYMENT_PER_CALL=0.50
MAX_DAILY_SPEND=10.00
You only need the key for chains you want to pay on. Solana-only? Just set SOLANA_PRIVATE_KEY.
3. Connect to Your Agent
Hermes Agent
hermes config set mcp_servers.x402.command "node"
hermes config set mcp_servers.x402.args '["/path/to/x402-agent-mcp/dist/index.js"]'
hermes config set mcp_servers.x402.enabled true
# Set env vars
python3 -c "
import yaml
with open('$HOME/.hermes/config.yaml') as f:
config = yaml.safe_load(f)
config['mcp_servers']['x402']['env'] = {
'SOLANA_PRIVATE_KEY': 'your-base58-key',
'EVM_PRIVATE_KEY': 'your-hex-key',
}
with open('$HOME/.hermes/config.yaml', 'w') as f:
yaml.dump(config, f, default_flow_style=False, allow_unicode=True)
"
hermes gateway restart
hermes mcp test x402
Claude Desktop
{
"mcpServers": {
"x402": {
"command": "node",
"args": ["/path/to/x402-agent-mcp/dist/index.js"],
"env": {
"SOLANA_PRIVATE_KEY": "your-base58-key",
"EVM_PRIVATE_KEY": "your-hex-key"
}
}
}
}
Usage Examples
Search for endpoints
x402_search({ query: "news" })
x402_search({ category: "social" })
x402_search({ chain: "solana" })
Discover a service by URL
x402_discover_url({ url: "https://svm402.com" })
Batch discover multiple URLs
x402_discover_urls({ urls: ["https://svm402.com", "https://2s.io"] })
Crawl x402scan for new services
x402_crawl_directory({ max_results: 20 })
Check if a service is live
x402_health({ name: "svm402" })
x402_health({ url: "https://svm402.com" })
Fetch an x402 endpoint
x402_fetch({
url: "https://svm402.com/analyze",
method: "POST",
body: '{"address": "DezXAZ8z7PnrnRJjz3wXBoRgixCa6xjnB7YaB1pPB263"}'
})
x402_fetch({
url: "https://2s.io/api/news/search?q=x402&limit=3",
chain: "solana"
})
Endpoint Directory
The endpoints.json file contains known x402 endpoints. It is gitignored — each installation builds its own directory.
endpoints.example.jsonis shipped as a template (empty, with categories)- On first run, the MCP loads the template and populates from there
x402_discover_urlauto-adds new services when discoveredx402_crawl_directoryscrapes x402scan.com for new services- Only true x402 endpoints (no API keys) are included
To bootstrap a fresh install:
cp endpoints.example.json endpoints.json
# Then run x402_crawl_directory to populate
Automated Directory Refresh
The endpoint directory stays fresh by running x402_crawl_directory on a schedule. Here's how to set it up in popular agent frameworks:
Hermes Agent (cron job)
Hermes supports scheduled cron jobs that can run the crawler automatically:
# Create a weekly cron job (every Monday at 03:00 UTC)
hermes cron create \
--name "x402 Crawl Directory" \
--schedule "0 3 * * 1" \
--toolsets terminal \
--deliver local \
--prompt 'Run the x402-agent-mcp crawler to discover new endpoints from x402scan.com:
cd /path/to/x402-agent-mcp && echo "{\"jsonrpc\":\"2.0\",\"id\":1,\"method\":\"initialize\",\"params\":{\"protocolVersion\":\"2024-11-05\",\"capabilities\":{},\"clientInfo\":{\"name\":\"hermes-cron\",\"version\":\"1.0\"}}}
{\"jsonrpc\":\"2.0\",\"method\":\"notifications/initialized\"}
{\"jsonrpc\":\"2.0\",\"id\":2,\"method\":\"tools/call\",\"params\":{\"name\":\"x402_crawl_directory\",\"arguments\":{\"max_results\":20}}}" | timeout 120 node dist/index.js 2>/dev/null | tail -1
Parse the JSON response and report how many new services were added. If none found, say "No new x402 endpoints discovered this week."'
The --deliver local flag keeps the cron job silent (no chat messages) — it just updates endpoints.json in the background.
Other Agents (crontab)
For agents without built-in scheduling, use system crontab:
# Add to crontab — runs every Monday at 03:00
crontab -e
# Add this line:
0 3 * * 1 cd /path/to/x402-agent-mcp && echo '{"jsonrpc":"2.0","id":1,"method":"initialize","params":{"protocolVersion":"2024-11-05","capabilities":{},"clientInfo":{"name":"cron","version":"1.0"}}}
{"jsonrpc":"2.0","method":"notifications/initialized"}
{"jsonrpc":"2.0","id":2,"method":"tools/call","params":{"name":"x402_crawl_directory","arguments":{"max_results":20}}}' | timeout 120 node dist/index.js >> /var/log/x402-crawl.log 2>&1
Custom Scripts
The crawler can also be called programmatically:
import { registerCrawlX402ScanTool } from "./tools/crawl-directory.js";
// Or call the MCP via stdio — see usage examples above
Disclaimer
This software is experimental and provided "as is", without warranty of any kind. Use at your own risk.
This software initiates real cryptocurrency transactions that are irreversible.
Tech Stack
- MCP SDK
- @x402/fetch — x402 payment handling
- @x402/svm — Solana x402 scheme
- @x402/evm — Base/EVM x402 scheme
- viem — EVM account signing
- @solana/kit — Solana SDK
License
MIT
Links
- x402 Protocol — Payment standard
- x402scan — Endpoint explorer
- MCP Protocol — Agent tool protocol
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.