rtl business documents pro

The problem

Default document generators produce broken Persian and Arabic layouts. You end up with disconnected letters, Western digits in the middle of RTL sentences, scrambled word order in mixed-language text, and tables flowing from the wrong side.

What it does

• Normalizes RTL strings using a deterministic Python engine to fix ZWNJ (half-spaces), digit conversion, and punctuation marks.
• Generates .docx and .pptx files using tested XML recipes that correctly set bidi flags and complex-script font slots.
• Embeds the open-license Vazirmatn font to ensure typography displays correctly across different PDF viewers and OSs.
• Handles mixed-direction text (LTR within RTL) using appropriate direction marks to prevent character reordering.
• Automates formal business document structure including honorifics, letterheads, and RTL-compliant table layouts.

Frameworks & tools

Python (rtl_normalize script), OpenXML (docx/pptx), and any PDF generation engine. Bundles Vazirmatn (SIL Open Font License).

Why this beats prompting it yourself

Prompting alone cannot fix underlying XML structure or binary encoding issues in document formats. This skill uses a Python-based normalization library and specific XML recipes to solve technical RTL rendering failures that standard LLM output ignores.

Use cases

• Generating formal Persian or Arabic business proposals and contracts with legal-grade typography.
• Creating localized pitch decks in PowerPoint with correct RTL slide transitions and table flows.
• Automating Persian invoice generation as PDFs with embedded fonts for consistent printing.
• Fixing broken "scrambled" Arabic text in existing technical documentation or reports.

Known limitations

Advanced normalization is optimized for Persian and Arabic. While basic RTL directionality works for Urdu or Hebrew, language-specific orthography rules for those languages are not included. Excel support is limited to string normalization and sheet-level RTL flags.