Japan Data Toolkit — addresses, names, banks, holidays (MCP)

    1

    Stop breaking on Japanese data: rules + a hosted MCP/API backed by official registries, free daily quota, no API key

    $9

    Secure checkout via Stripe

    30-day refund guarantee

    Converts to your local currency at checkout

    0 installsSecurity scanned

    Works with the AI tools you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIGitHub CopilotGitHub CopilotGemini CLIGemini CLI+20 more

    Japan Data Toolkit — addresses, names, banks, holidays (MCP)

    Example session with this skill installed

    Here are 3 rows from our Japanese customer export. Clean them up for our Shopify import: split names, add romaji for the shipping label, parse the address into fields with lat/lng, and give me the payment due date if we invoice on 2026-09-18 with net 3 business days (bank calendar).

    氏名,フリガナ,住所,電話
    佐藤裕子,さとう ひろこ,東京都千代田区丸の内１丁目１－１ 東京ビル3F,０３－１２３４－５６７８
    大野遼,,東京都千代田区千代田１−１,03 1234 5678
    小島裕子,,〒100-0005 千代田区丸の内1-1-1,+81-3-1234-5678

    • Read your context and instructions
    • Compiled the japan data toolkit
    • Generated the data export

    Summary

    3 rows processed with jp_text_normalize → jp_name_parse (フリガナ passed as kana when present) → jp_address_normalize → jp_business_day.

    1 row needs human review (given-name reading ambiguous, no フリガナ). Address geocoding resolved all 3 to town level (level 3); block-level (level 8) needs the 号 written out.

    Cleaned rows

    #familygivenreadingpassport romajilabel (Western)prefcitytownrestlat / lngphone (E.164)review
    1佐藤裕子さとう ひろこ (supplied)SATO HIROKOHiroko Sato東京都千代田区丸の内一丁目1-1 東京ビル3F35.68156 / 139.767201+81312345678–
    2大野遼おおの りょう (0.86) / はるか (0.78)ONO RYORyo Ono東京都千代田区千代田1-135.684052 / 139.752145+81312345678given-name reading: medium
    3小島裕子こじま ゆうこ (0.86) / ひろこ (0.86) / のぶこ / のりこKOJIMA YUKOYuko Kojima東京都千代田区丸の内一丁目1-135.68156 / 139.767201+81312345678given-name reading: tie — ask customer

    Notes applied automatically

    • Full-width digits/hyphens → half-width; 〒100-0005 stripped into a separate postal column (1000005).
    • Row 3 had no prefecture; the registry resolved 千代田区 → 東京都 unambiguously.
    • Passport romaji follows 外務省 Hepburn: long vowels dropped (佐藤 → SATO, 裕子 → YUKO; 大野 → ONO, not OHNO). If the customer's passport uses OH-style, pass ohStyle=true to jp_name_romaji.

    Payment due date

    jp_business_day?date=2026-09-18&add=3&calendar=bank →

    { "date": "2026-09-18", "weekday": "Friday", "isBusinessDay": true,
  "nextBusinessDay": "2026-09-24", "add": 3, "result": "2026-09-28",
  "lastBusinessDayOfMonth": "2026-09-30" }

    Due 2026-09-28 (Mon). 3 business days is not Sep 23: Sep 19–23 is a 5-day break (Sat, Sun, 敬老の日 Mon 21, 国民の休日 Tue 22, 秋分の日 Wed 23). Month-end cutoff for 月末締め would be Sep 30.

    Shopify-ready CSV (first row)

    First Name,Last Name,Company,Address1,Address2,City,Province,Zip,Country,Phone,Note
Hiroko,Sato,,丸の内一丁目 1-1,東京ビル3F,千代田区,東京都,,JP,+81312345678,reading=さとう ひろこ; geo=35.68156,139.767201

    Next steps

    1. Confirm row 3's reading with the customer (ゆうこ vs ひろこ) before printing the label — both score 0.86.
    2. For block-level coordinates (level 8) ask customers for the full 丁目-番-号; the registry matched only up to the town here.
    3. Run the whole export with node scripts/clean-customers.mjs customers.csv > cleaned.csv — same pipeline, adds confidence/level/review columns so only low-confidence rows go to a human.

    japan-data-toolkit-addresses-names-banks.csv

    CSV · data export

    Generated

    Example file from a real run - the skill writes it into your workspace.

    Connects securely to your tools. The creator never sees your data.

    What you get

    Convert Japanese names to passport-compliant Hepburn Romaji for shipping.Resolve Japanese bank and branch codes for Zengin payment files.Normalize messy Japanese addresses into structured prefecture/city/block fields.Check Corporate Numbers (法人番号) against the National Tax Agency registry.

    About this skill

    The problem

    Japanese business data uses complex formatting, inconsistent address notation, and ambiguous character readings that break standard ETL pipelines. Manually handling full-width character normalization, Japanese era dates (Wareki), and official registry lookups is error-prone and time-consuming.

    What it does

    • Normalizes Japanese text by converting full-width/half-width characters and extracting dates, phone numbers, and postal codes.
    • Validates and parses Japanese addresses down to the block level using official registries, providing WGS84 coordinates.
    • Resolves Japanese bank and branch codes while generating the specific half-width kana required for Zengin transfer files.
    • Performs KYC lookups against the National Tax Agency registry to retrieve official Corporate Numbers (法人番号) and registered addresses.
    • Calculates Japanese business days and bank holidays, including substitute holidays and complex year-end closures.

    Frameworks & tools

    Accessible via MCP server (Claude Desktop, Cursor, Claude Code) or REST API. Includes an MCP configuration for ja-normalize and scripts for Node.js environments.

    Why this beats prompting it yourself

    Standard LLMs struggle with deterministic Japanese data tasks like passport-compliant Hepburn romanization or resolving exact bank branch codes. This skill connects your agent to live, registry-backed tools that eliminate hallucinations in critical fields like financial routing or legal identifiers.

    Use cases

    • Clean and de-duplicate Japanese customer CRM records with verified address components.
    • Automate accounts payable by resolving bank codes and calculating valid bank settlement dates.
    • Prepare international shipping labels by converting Japanese names to official passport-grade Romaji.
    • Verify corporate existence and retrieve official registration numbers for Japanese suppliers.

    Known limitations

    Free tier is limited to 20 calls per IP per day without an X402 payment key. Address resolution accuracy depends on the Digital Agency Address Base Registry coverage.

    How to install

    Works the same in every agent - Claude, Cursor, Codex, Copilot and 20+ more.

    ~30 seconds
    1. 1

      Download the ZIP

      Free skills download straight away. Paid skills unlock right after purchase.

    2. 2

      Unzip into your skills folder

      Every agent reads skills from one folder on your machine. Drop the unzipped folder in there.

    3. 3

      Ask your agent to use it

      Restart the agent if it was already running. It picks the skill up automatically - no config needed.

    Skills folder by agent

    Click the path to copy it. Create the folder if it does not exist yet.

    Reviews

    No reviews yet

    Be one of the first to try it. Every listed skill passes our trust checks below.

    Security scanned

    Passed our 8-point scan before listing

    Fresh listing

    Recently published to Agensi

    30-day refund

    Not a fit? Get your money back

    Trust & safety

    Security scanned

    Verified clean 20 days ago

    • Passed all security checks, Safe to install

    Listed20 days ago

    What's inside

    Frequently Asked Questions