Japan Data MCP server

japan-data-mcp

npm: @mameta/japan-data-mcp · MCP Registry: io.github.Mameta29/japan-data-mcp · transport: stdio

An MCP server that gives AI agents (Claude Desktop, Cursor, Cline, …) one-call access to the agentic-jp.com suite of Japanese data APIs. Every call is settled per-use over the x402 payment protocol — no account, no subscription, no API key.

What it covers — 28 tools across 7 APIs

How payment works

x402 is a buyer-pays protocol. This MCP server holds your wallet and pays each API call transparently — the agent never sees a payment prompt. Calls cost roughly $0.001–$0.10 each (mostly sub-cent), settled in USDC on Base.

• You fund a wallet with a small amount of USDC.
• You set EVMPRIVATEKEY to that wallet's key.
• The server pays as it goes; per-call cost is well under a cent for most tools.

Without EVMPRIVATEKEY the server still starts and lists every tool, but a paid tool call returns a clear "payment not configured" message — nothing is charged.

Security: the key in EVMPRIVATEKEY can spend that wallet's funds. Use a

dedicated wallet funded with only what you intend to spend — never a primary

wallet.

Setup

No install step — npx fetches and runs the server on demand.

Claude Desktop

Add to claudedesktopconfig.json:

{
  "mcpServers": {
    "japan-data": {
      "command": "npx",
      "args": ["-y", "@mameta/japan-data-mcp"],
      "env": {
        "EVM_PRIVATE_KEY": "0xYOUR_FUNDED_WALLET_KEY"
      }
    }
  }
}

Without EVMPRIVATEKEY the server still starts and lists all 28 tools; the paid tools then return a clear "payments disabled" error instead of running. Add a USDC-funded wallet key to enable them.

Cursor / Cline

Point the MCP client at npx -y @mameta/japan-data-mcp with the same EVMPRIVATEKEY env var. Transport is stdio.

From source

git clone https://github.com/Mameta29/japan-data-mcp.git
cd japan-data-mcp
npm install && npm run build
# then point the MCP client at: node /abs/path/to/dist/index.js

Example prompts

Once connected, just ask the agent in natural language:

Shortened. The full README is on GitHub.