Skip to main content

nexus-ja-data

A local MCP server that lets AI agents clean Japanese business data with fixed rules instead of guessing.

  • Normalizes Japanese company names ((株) / ㈱ / 株式会社 / half-width kana), phone numbers, postal codes, emails, addresses and era dates (令和6年4月1日, R6.3.15 → 2024-04-01)
  • Deduplicates Japanese customer lists (名寄せ). Flags similar-but-different companies (さくら商事 / さくら商会) for review instead of merging them
  • Same input → same output. Returns the reason for each match
  • Python standard library only. No network access. Data never leaves your machine
  • MIT license
{ "mcpServers": { "nexus-ja-data": { "command": "python", "args": ["/path/to/nexus-ja-data/run_server.py"] } } }

Tools: ja_normalize_value, ja_same_company, ja_detect_column_types, ja_clean_and_dedupe_records. See llms.txt.


nexus-ja-data(日本語)

AIエージェントのための、日本語ビジネスデータ整形・名寄せの道具(MCPサーバー)。 ClaudeやChatGPTなどのAIに登録すると、AIが日本語の会社名・電話・住所・和暦を「推測」ではなく「決まったルール」で正確に直し、顧客名簿の重複をまとめられるようになる。

  • Python標準ライブラリだけで動く(追加インストール不要)
  • 外部通信なし。データはパソコンの外に出ない
  • 名寄せ赤ペン(p004)と同じルール。JavaScript版と結果が一致することをテストで確認済み

AIが使える道具(4つ)

名前 できること
ja_normalize_value 値1つを種類に合わせて直す(会社名・氏名・フリガナ・電話・郵便番号・メール・住所・日付)
ja_same_company 2つの会社名が同じ会社かを判定((株)/株式会社/半角カナ/ひらがな の違いを吸収、似ている別会社は「別」)
ja_detect_column_types 表の列の種類を推定
ja_clean_and_dedupe_records 名簿を正規化して重複をまとめる(名寄せ)

なぜAIに必要か

大きなAIでも、日本語の表記ゆれは毎回少しずつ違う直し方をしたり、「さくら商事」と「さくら商会」を同じ会社にしてしまったりする。この道具を使えば、同じ入力にいつも同じ答えを返し、判断の根拠(一致した条件・似ている度)も返す。

使い方(Claude Desktop の例)

設定ファイル claude_desktop_config.json に次を追加(パスは置いた場所に合わせる):

{
  "mcpServers": {
    "nexus-ja-data": {
      "command": "python",
      "args": ["-m", "nexus_ja.server"],
      "cwd": "C:\\Users\\あなた\\Documents\\nexus-ja-data"
    }
  }
}

cwd が使えないアプリでは "args": ["C:\\...\\nexus-ja-data\\run_server.py"] を指定する。

配布条件

MITライセンス(無料・商用利用可)。行数の制限なし。 人間が画面で使える同じルールの道具「名寄せ赤ペン」は BOOTH で販売中: https://nexus-tools.booth.pm/items/9051166

品質

  • テスト11件(python -m unittest discover -s tests): 正規化の期待値、名寄せ、MCPとしての通し会話(初期化→一覧→呼び出し→エラー処理)、壊れた入力でも落ちない、外部通信コードなし、JavaScript版との一致
  • 2026-10-27: 全件成功

Metadata

Release files for nexus-ja-data 0.2.0

For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.

Source distribution (sdist)

Source distribution for nexus-ja-data 0.2.0
File Size Uploaded Provenance
nexus_ja_data-0.2.0.tar.gz 19.3 kB Without provenance Details

Built distribution (wheel)

Table of built distributions (wheels) for nexus-ja-data 0.2.0
File Interpreter ABI Platform
nexus_ja_data-0.2.0-py3-none-any.whl Python 3 none any Details

Total release size: 34.6 kB

Release files / nexus_ja_data-0.2.0.tar.gz

Download URL nexus_ja_data-0.2.0.tar.gz
Size 19.3 kB
Tags Source
SHA-256 checksum
How to use checksums
9caf72807486a40785cf452441d393d2a29efee3973f02420a8ab674c3777257
BLAKE2b-256 checksum
How to use checksums
06cae803c1383a01c1a73e333d68fde84465bce4690b1fbe125a8cd16771ccea
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
No
Uploaded via twine/7.0.0 CPython/3.14.7

Release files / nexus_ja_data-0.2.0-py3-none-any.whl

Download URL nexus_ja_data-0.2.0-py3-none-any.whl
Size 15.3 kB
Tags Python 3
SHA-256 checksum
How to use checksums
9cc372703ccc645f8d7d13d9aeccef823cd9c50a9408b9c86bfa65b488b9e938
BLAKE2b-256 checksum
How to use checksums
f205f8a396f5e8e6d5128e480905d76dc73da4b83883f9d4da27ed63a0686a9f
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
No
Uploaded via twine/7.0.0 CPython/3.14.7

Security for nexus-ja-data 0.2.0

Provenance

Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.

No provenance found

None of the files in this release were uploaded with provenance metadata.

A note on package security

PyPI is a repository for distributing packages, not a service for auditing them. While third-party security researchers and our staff do their best to detect and remove malicious packages, PyPI cannot guarantee the quality, security or suitability of any given package.

To learn more about protecting yourself from malicious actors, visit our package security documentation.

Release history Release notifications | RSS feed

This release

0.2.0 This release

2 release files No provenance

Anthropic, PBC Visionary sponsor Bloomberg Visionary sponsor Hudson River Trading Visionary sponsor Meta Visionary sponsor NVIDIA Visionary sponsor Microsoft Sustainability sponsor Depot Continuous Integration AWS Cloud computing and Security Sponsor Datadog Monitoring Fastly CDN Google Download Analytics Sentry Error logging StatusPage Status page