Word Aligner

io.github.tinygodsdev由 GitHub 帳號 tinygodsdev 發布v1.0.0更新於 Oct 24, 2026

Turn a phrase and its translation into a shareable word-alignment diagram.

已驗證Streamable HTTP可網頁執行知識與記憶媒體與設計

概覽

AI 產生的概覽

一個遠端 MCP 伺服器，可將兩行或多行堆疊文字產生可分享的詞對齊示意圖，並回傳連結與預覽圖。

可以做什麼
依堆疊的多行文字（例如譯文與原文、註解或 IPA）建立詞對齊示意圖，標示各行之間哪些詞彼此對應，並回傳可分享的連結與預覽圖。
適用情境
適合需要呈現詞組與譯文逐詞對應關係、製作行間註解，或為課程、文法筆記、人造語言文件產生可分享圖示的情境。
執行需求
遠端 Streamable HTTP 端點；未宣告需要身分驗證、環境變數或請求標頭。README 說明其無需驗證即可使用。
安裝前請注意
提交的文字會傳送到第三方服務並編碼進可分享連結，請避免傳送敏感或私密內容。此工具只產生連結與預覽圖，不會修改既有資料。

安裝

在 SourceWeft 中

  1. 開啟 儀表板中的 Word Aligner，將其新增到工作區。
  2. 為需要使用其工具的對話啟用該服務。

Web executable，透過 Streamable HTTP。 遠端服務在工作區中設定後即可從網頁執行環境執行。

交給代理程式安裝

把這段提示詞貼上到 Claude Code、Codex、Cursor 等代理程式，或 SourceWeft 對話中。代理程式會讀取該伺服器的安裝指南，先向你展示發布者、端點和登入方式，經你同意後再把它新增到你的客戶端。金鑰不會寫進任何命令。

閱讀 https://sourceweft.com/mcp/io.github.tinygodsdev%2Fword-aligner/install.md，並按其中說明把這個 MCP 伺服器接入我的客戶端。修改任何設定之前，先告訴我它的發布者、端點和登入方式，等我確認。不要把 API 金鑰寫進命令或我會提交的檔案；問我金鑰應從哪裡取得。

檢視代理程式所依據的安裝指南

其他 MCP 客戶端

把它新增到你客戶端的 mcpServers 設定中。

{
  "mcpServers": {
    "word-aligner": {
      "type": "http",
      "url": "https://aligner.tinygods.dev/mcp"
    }
  }
}

README

Aligner — word-by-word translation visualizer

aligner.tinygods.dev

Word Aligner is a free web tool that shows which word matches which across stacked lines of text. Type or paste a sentence and its translation, click a word and then its match on the line above or below, and curved connectors draw the alignment between them. Add extra rows for glosses or IPA, then export the diagram or share it with a link.

No accounts, no machine translation — you stay in control of every link. Great for lessons, social posts, grammar notes, and conlang documentation.

[Linking words between “Hello world” and its French translation]

What it's for

  • Language learners — see exactly why a translation says what it does: which source word became which target word, what gets dropped, and where word order changes.
  • Teachers — build handouts, slides, and flashcards; export a clean PNG or SVG and drop it into a lesson without retyping anything.
  • Linguists — produce Leipzig-style interlinear glosses with morpheme alignment, case/agreement tags, and free translations.
  • Conlangers — show how a constructed language maps onto English, with custom script fonts, gloss rows, and IPA.

What it does

  • Multi-line alignment — stack as many rows as you need (source, translation, gloss, IPA); links connect vertically adjacent lines.
  • Manual word linking — click a word, then its match; supports one-to-many and many-to-one links and freely crossing connectors for reordered translations.
  • Interlinear glosses & IPA — add annotation tiers above or below any line.
  • Right-to-left scripts — Hebrew, Arabic, and mixed LTR/RTL layouts.
  • Vertical writing systems — stand the whole diagram up as columns for Japanese tategaki, Chinese, or traditional Mongolian; characters stack or the line rotates, chosen per line, so a horizontal translation stays readable beside it.
  • Tokenization control — choose how text splits into word boxes (split characters, a join marker for fixed expressions, optional punctuation handling).
  • Per-line typography — font, size, spacing, and direction per line; custom and Google Fonts.
  • Exports — PNG, SVG, PDF, and a self-contained HTML file; exports match the preview, including custom-font shaping.
  • Shareable URLs — the whole alignment (text, links, and visual settings) is encoded in a ?data= link.

Examples

Browse ready-made examples — bilingual pairs, Turkish interlinear with IPA, RTL scripts, vertical Japanese and traditional Mongolian, Tagalog compounds, Japanese–Chinese–English word order, and more interlinear glosses — at aligner.tinygods.dev/examples. Open any one in the editor to adapt it.

[Turkish interlinear preview with glosses, IPA, and connectors]

[Conlang example with a custom script font and interlinear glosses]

How it works

Edit your lines in the line editor, then link words in the preview.

[Aligner line editor with stacked lines and word-splitting summary]

Tune colors, tokenization, and fonts in settings, then export or share.

[Export options: PNG, SVG, PDF, and HTML]

Word alignment vs. interlinear vs. parallel text

  • Interlinear translation places a gloss directly under each source word — compact, but it hides reordering. Aligner keeps both sentences on their own line and draws connectors, so reorderings, splits, and merges stay obvious.
  • Parallel text is side-by-side bilingual reading for long-form study. Aligner focuses on one sentence pair at a time with explicit connectors showing which tokens correspond.

API

Word Aligner has a free HTTP API for generating alignment links programmatically.

POST https://aligner.tinygods.dev/api/alignContent-Type: application/json
{  "lines": ["Hello world", "Bonjour le monde"],  "alignments": [[0, 0, 1, 0], [0, 1, 1, 2]]}

Returns { "url": "https://aligner.tinygods.dev/?data=..." }.

Supports per-line options (font, size, RTL), global visual settings (palette, line style, background), pair-level gap control, and multi-line stacks with selective connectors.

Using Word Aligner as an AI agent tool

The word-aligner-skill/ directory is a ready-to-install skill that teaches AI assistants (Claude, ChatGPT, etc.) how to use the API. Once installed, it enables flows like:

"Translate 'я хочу спать' into English and show me the word alignment."

The agent translates, calls the API, and returns a shareable link.

  • For Claude: point your Claude project at this repository or add the contents of word-aligner-skill/SKILL.md as a custom instruction.
  • For ChatGPT: zip word-aligner-skill/ and upload it as a custom Skill.

Install instructions and a downloadable archive are also on the site: aligner.tinygods.dev/skill.

MCP server

There is also an MCP server at https://aligner.tinygods.dev/mcp that exposes a single create_word_alignment tool. It returns a shareable link plus a preview image, and runs inside the same app over Streamable HTTP with no auth. Connect it in Claude:

bash
claude mcp add --transport http word-aligner https://aligner.tinygods.dev/mcp

See docs/mcp-server.md for ChatGPT Developer Mode setup and the tool schema.

Learn more

Created by Dani Polani. See other tools for linguistics and conlanging.

來源：README.md，提交 c794a8b

工具

1

create_word_alignment

Create word alignment diagram

Create a shareable Word Aligner diagram that shows which words match across two or more stacked lines of text (a translation and its source, an interlinear gloss, IPA, etc.). Returns a URL that opens the interactive diagram, plus a preview image. Use this when the user wants to translate a phrase and show word correspondences, align a translation with its source (including RTL scripts like Hebrew or Arabic, or vertically written ones like Japanese and Mongolian), or build a Leipzig-style interlinear gloss. Word indices are 0-based token positions. Tokenize each line the same way the tool does before assigning indices: - Whitespace always splits ("I have been going" -> I[0] have[1] been[2] going[3]). - The characters in settings.tokenSplitChars (default ".-|") also split and are then removed from the rendered text, so "go.PST.IPFV" becomes three tokens (go, PST, IPFV) and the dots disappear. For Leipzig glosses set tokenSplitChars to "-|" to keep the dots. - Punctuation stays attached by default ("Hello, world!" -> Hello,[0] world![1]). - In RTL lines, word 0 is the logically first word (rightmost on screen); index in reading order. - Japanese and Chinese are written without spaces and nothing is segmented for you: put spaces where the alignment units should be. For a vertically written script set settings.axis to "columns". Every line then becomes a vertical column and the connectors run sideways. Set orientation per line: "vertical" stacks the characters (Japanese, Chinese), "sideways" rotates the line a quarter turn (traditional Mongolian, and Latin runs inside vertical text), "upright" leaves a translation as horizontal word boxes. The first line is the leftmost column, so for Japanese and Chinese, whose columns read right to left, list the translation first and the script second. Each alignment is [lineA, wordA, lineB, wordB]; the two lines must be neighbours in the stack (|lineA - lineB| = 1), which means one above the other in rows and side by side in colu

輸入參數lines必填pairssettingsalignments

版本歷史

1
  1. v1.0.0最新Sep 16, 2026

評分

已滿足適用檢查項的 100%

該評分衡量此收錄條目的完整程度，依據註冊中心條目、端點與公開來源計算得出。它不是對伺服器本身的評價，也不表示伺服器安全或目前可用：可連通性是帶日期的檢查結果，評分也不包含安裝量或使用者評價。

  • 近 30 天內可連通必需25 分
  • 已宣告認證方式必需10 分
  • 已列出工具必需15 分
  • 描述必需15 分
  • 圖示10 分
  • 文件15 分
  • 授權條款10 分