FIRST CH TOOLS LOCAL PROCESSING — NOTHING IS UPLOADED 日本語

FIRST CH TOOLS / SEO & AI / 66 URL SLUG GENERATOR

URL Slug Generator (Japanese → Romaji)

Turns a Japanese article title into a URL slug for WordPress and similar CMSs. Kana are written in Hepburn romaji, symbols are removed, and the words are joined with hyphens and lowercased. For kanji, type the reading in kana in the reading field and the slug is built from that. Nothing you type leaves your device.

1 — Title and reading

Leave it empty to build the slug from the title. Anything in this field is used instead of the title. Each space becomes a separator (hyphen), so put spaces between words. Latin letters and digits can be used as they are.
Separator

2 — Slug

https://example.com/…/

Romaji: —

0
Characters
0
Words
title
Built from
0
Skipped characters (kanji etc.)
Word by word (reading → romaji)
Other spellings (click to copy)
Checks

    A title typed on screen never leaves your browser (the conversion runs entirely on this page). You can also pass it in the URL (values in the URL reach the server as part of the page request): /en/slug/?title=はじめてのワードプレス / /en/slug/?reading=とうきょう%20らーめん&sep=snake&long=keep

    How to Use

    1. Paste the titlePaste the article or page title as it is. If it contains only kana, katakana and Latin letters or digits, the slug appears straight away.
    2. Fix kanji in the reading fieldPress "Copy title to reading" and rewrite only the kanji in kana. Each space becomes a hyphen, and deleting particles or filler words shortens the URL.
    3. Copy itUse "Copy" next to the slug and paste it into the URL slug (permalink) field of the WordPress editor. Underscore and spelled-out long-vowel versions can be copied in one click under "Other spellings".

    About This Tool

    A Japanese title used as a slug as it is turns the URL into a long string such as %E6%9D%B1%E4%BA%AC…. It is unreadable when pasted into social media or chat and adds a lot of length, so keeping the slug in Latin letters and digits is the usual practice. This tool does that conversion by fixed rules, on your device.

    The romaji follow Hepburn. し → shi, ち → chi, つ → tsu, ふ → fu, じ → ji; contracted sounds are しゃ → sha and ちゃ → cha; a small っ doubles the next consonant, きって → kitte (and まっちゃ → matcha before ch); ん is always n (しんぶん → shinbun). Loanword sounds such as ファ → fa, ティ → ti and ヴァ → va are covered. The particles は・へ・を are written as spelled, ha / he / o; rewrite them in the reading field if you want “wa” or “e”.

    Long vowels are dropped by default. とうきょう → tokyo, おおさか → osaka, ラーメン → ramen: “ou”, “oo”, “uu” and the long-vowel mark ー collapse to one letter, as on passports and station signs (せんせい → sensei and いいね → iine stay as they are). “Spell out” gives toukyou and raamen. What matters most is using one style across the site.

    Kanji readings are not guessed. Deciding how kanji are read takes a morphological dictionary of tens of megabytes, and even then names, places and proper nouns often come out wrong. Because a URL is hard to change once published, a person decides the reading and types it in the reading field. If kanji remain with the reading field empty, they are skipped, the slug is built from the kana, letters and digits only, and the skipped kanji are listed under Checks.

    Words are split at spaces, symbols, kanji and changes between kana and Latin letters or digits (ラーメン10選 → ramen-10). By default a change between hiragana and katakana also splits, so katakana loanwords become separate words. Half-width katakana and full-width letters are normalised first, and accents are stripped (Café → cafe). The result contains only a–z, 0–9 and the separator, so WordPress’s sanitize_title leaves it unchanged.

    A digits-only slug (such as 2026) can clash with date archives depending on your permalink settings, and WordPress may save it with -2 appended, so it is flagged under Checks — as are slugs longer than 60 characters. For how a page looks when shared, see the OGP Meta Tag Wizard; for search results, the SERP Preview; for identifier case conversion, the Case Converter.

    From AI Agents

    The same slug logic is available as the slug_generate tool of the MCP (Model Context Protocol) server @first-ch/tools-mcp, so an AI agent can call it directly without driving a browser. Let the AI decide the reading and leave the romanisation, symbol removal and separator rules to this tool. See How to use from AI agents for the details.

    Install

    claude mcp add firstch-tools -- npx -y @first-ch/tools-mcp

    Examples

    # Build a slug from a reading (the caller decides how the kanji are read)
slug_generate(title="東京で食べたラーメン10選", reading="とうきょう で たべた ラーメン 10せん")

# Underscore separator, long vowels spelled out
slug_generate(reading="こうしき ブログ", separator="snake", longVowel="keep")

# Many article titles at once (up to 1000)
slug_generate(items=[{"title": "はじめてのワードプレス"}, {"title": "SEO対策", "reading": "SEO たいさく"}], maxLength=50)

    Other Tools