arabic-text-repair

name	arabic-text-repair
description	Skill for detecting, repairing, and standardizing Arabic text typography, fixing inverted punctuation, Tatweel/Kashida normalization, digit formatting (Eastern/Western), and Alef/Yeh forms.

Arabic Text Repair & Typography Skill

This skill guides the automated detection, diagnosis, and precision repair of Arabic text defects and formatting inconsistencies before insertion into Microsoft Word documents.

---

1. Common Defects & Diagnostic Criteria

1. Inverted Parentheses & Punctuation Marks:

   • In mixed Arabic/Latin contexts or numeric parentheticals, punctuation marks such as (, ), [, ], . frequently flip orientation in Word.
   • Solution: Set fixInvertedPunctuation: true.

2. Digit Format Inconsistency:

   • Documents mixing Western numerals (1, 2, 3) with Eastern Arabic numerals (١, ٢, ٣).
   • Solution: Standardize across the document via standardizeDigits: 'eastern' (or 'western').

3. Alef & Dotless Yeh Standardization:

   • Unifying hamza forms on Alef (أ, إ, آ -> ا) or normalizing final Yeh (ى -> ي).
   • Solution: Enable normalizeAlef: true and/or normalizeYeh: true.

4. Excess Kashida (Tatweel) Removal:

   • Stripping unwanted tatweel (ـ) characters from extracted OCR or legacy text.
   • Solution: Set removeKashida: true.

---

2. Execution via MCP Tool

{
  "tool": "repair_arabic_text_formatting",
  "arguments": {
    "text": "النص العربي المراد فحصه وإصلاحه (123) ـتـجـربـةـ",
    "fixInvertedPunctuation": true,
    "standardizeDigits": "eastern",
    "trimExtraSpaces": true,
    "normalizeAlef": false,
    "normalizeYeh": false,
    "removeKashida": true
  }
}