How do I install the Uae Arabic Documents skill?

Run npx skillmds@latest add habib-alqalam/uae-arabic-documents in your terminal (requires Node.js), paste this page's agent-chat prompt into Claude, Cursor, or any MCP-connected agent, or download the SKILL.md file and copy it into your agent's skills directory.

Is Uae Arabic Documents safe to use?

This skill has not completed SkillMD's automated safety review yet. SkillMD never runs a skill's scripts for you; review the SKILL.md before installing.

Which AI agents work with Uae Arabic Documents?

This skill is tagged as working with Claude Code, Claude.ai, OpenAI Codex. SKILL.md is an open format, so most agents that read a skills directory can load it too.

Is Uae Arabic Documents free to use?

Yes. Installing skills from SkillMD is free, and the skill stays under its author's original license.

Who published Uae Arabic Documents?

habib-alqalam (@habib-alqalam) published this skill. Their other Agent Skills are listed on their SkillMD profile.

Uae Arabic Documents by habib-alqalam

Name: Uae Arabic Documents
Author: habib-alqalam

Uae Arabic Documents

Generate UAE government and official Arabic documents (letters, memoranda to authorities, court filings, quotations, invoices) as correctly rendered PDFs — and fix any broken Arabic PDF output. ALWAYS use this skill when the task involves Arabic text in a PDF, RTL rendering, arabic_reshaper/python-bidi, reversed or disconnected Arabic letters, reversed line order in Arabic paragraphs, Arabic bullet points or numbered lists, tofu boxes, Emirates ID / TRN / trade licence / case numbers inside Arabic text, Hijri or dual dates, AED amounts in words (tafqit), bilingual AR/EN documents, or any letter/memo/quotation addressed to a UAE ministry, court, prosecution, free zone or government department. Also use it to DIAGNOSE broken Arabic output — the bug catalog maps every known failure mode to its fix.

habib-alqalam Updated Sep 17, 2026

File contents

uae-arabic-documents

Engineering standard + working code for generating UAE official Arabic documents that render correctly, plus an automated verification gate.

Workflow

Read references/rendering-rules.md first — the 10 rules. Every Arabic rendering decision in this skill follows them.
For UAE structure/register/numbering/dates/identifiers, read references/uae-conventions.md.
Diagnosing broken output? Go straight to references/bug-catalog.md — failure mode → root cause → fix.
Build with the engine (never raw reshape+bidi into a Paragraph):

from engine.ar_text import ar, iso                # single lines / tokens
from engine.rtl_paragraph import rtl_paragraphs   # multi-line paragraphs
from engine.rtl_lists import rtl_list             # bullets & numbering
from engine.fonts import check_font               # gate 1: font coverage
from engine.verify import verify_pdf              # gate 2: reading order
from uae.identifiers import emirates_id, trn, case_number, phone
from uae.dates import dual_date_ar, gregorian_ar
from uae.amounts import aed, tafqit_aed

Document skeletons: uae/templates/ (official letter, prosecution memorandum, VAT quotation) — {{variable-name}} placeholders.
End-to-end reference implementation: demo/generate_demo_letter.py.

Non-negotiable rules (full detail in references/)

ar() output is visual order → single lines only. Multi-line text goes through rtl_paragraphs() or lines will stack reversed.
Every embedded LTR token (numbers, refs, English) through iso() — or use the typed wrappers in uae/identifiers.py.
Directional controls are stripped after BiDi (ar() does this); left in, ReportLab silently drops everything after them.
from bidi import get_display (≥0.5) — never bidi.algorithm.
Fonts: verify Presentation Forms B coverage with check_font() before building. FreeSerif/Amiri/Noto yes; FreeSans/Helvetica no. Never italic, never synthetic bold, never fallback, TTF only.
Lists via rtl_list() — never bulletText=/ListFlowable.
Dates: written Gulf month names + Latin digits; dual-date line for official correspondence via dual_date_ar().
Amounts: isolated figure via aed() + mandatory tafqit line via tafqit_aed() on quotations/receipts/contracts.
Every generated PDF runs verify_pdf() before delivery. FAIL = fix and rebuild, never ship. PASS = proceed to human review.
Also render a page preview image and visually inspect it — the verifier is a gate before human review, not a replacement for it.

---name: uae-arabic-documentsdescription: Generate UAE government and official Arabic documents (letters, memoranda to authorities, court filings, quotations, invoices) as correctly rendered PDFs — and fix any broken Arabic PDF output. ALWAYS use this skill when the task involves Arabic text in a PDF, RTL rendering, arabic_reshaper/python-bidi, reversed or disconnected Arabic letters, reversed line order in Arabic paragraphs, Arabic bullet points or numbered lists, tofu boxes, Emirates ID / TRN / trade licence / case numbers inside Arabic text, Hijri or dual dates, AED amounts in words (tafqit), bilingual AR/EN documents, or any letter/memo/quotation addressed to a UAE ministry, court, prosecution, free zone or government department. Also use it to DIAGNOSE broken Arabic output — the bug catalog maps every known failure mode to its fix.---# uae-arabic-documentsEngineering standard + working code for generating UAE official Arabicdocuments that render correctly, plus an automated verification gate.## Workflow1. **Read `references/rendering-rules.md` first** — the 10 rules. Every Arabic rendering decision in this skill follows them.2. For UAE structure/register/numbering/dates/identifiers, read `references/uae-conventions.md`.3. Diagnosing broken output? Go straight to `references/bug-catalog.md` — failure mode → root cause → fix.4. Build with the engine (never raw reshape+bidi into a Paragraph):```pythonfrom engine.ar_text import ar, iso # single lines / tokensfrom engine.rtl_paragraph import rtl_paragraphs # multi-line paragraphsfrom engine.rtl_lists import rtl_list # bullets & numberingfrom engine.fonts import check_font # gate 1: font coveragefrom engine.verify import verify_pdf # gate 2: reading orderfrom uae.identifiers import emirates_id, trn, case_number, phonefrom uae.dates import dual_date_ar, gregorian_arfrom uae.amounts import aed, tafqit_aed```5. Document skeletons: `uae/templates/` (official letter, prosecution memorandum, VAT quotation) — `{{variable-name}}` placeholders.6. End-to-end reference implementation: `demo/generate_demo_letter.py`.## Non-negotiable rules (full detail in references/)- `ar()` output is visual order → **single lines only**. Multi-line text goes through `rtl_paragraphs()` or lines will stack reversed.- Every embedded LTR token (numbers, refs, English) through `iso()` — or use the typed wrappers in `uae/identifiers.py`.- Directional controls are stripped **after** BiDi (`ar()` does this); left in, ReportLab silently drops everything after them.- `from bidi import get_display` (≥0.5) — never `bidi.algorithm`.- Fonts: verify Presentation Forms B coverage with `check_font()` before building. FreeSerif/Amiri/Noto yes; FreeSans/Helvetica no. Never italic, never synthetic bold, never fallback, TTF only.- Lists via `rtl_list()` — never `bulletText=`/`ListFlowable`.- Dates: written Gulf month names + Latin digits; dual-date line for official correspondence via `dual_date_ar()`.- Amounts: isolated figure via `aed()` + mandatory tafqit line via `tafqit_aed()` on quotations/receipts/contracts.- **Every generated PDF runs `verify_pdf()` before delivery.** FAIL = fix and rebuild, never ship. PASS = proceed to human review.- Also render a page preview image and visually inspect it — the verifier is a gate before human review, not a replacement for it.

Frequently asked questions