Correctly shaped right-to-left PDFs for Arabic, Urdu and Persian text.
Maintain this server? Claim it
Call this server's tools from your browser. Nothing to install.
Requests go through mcptry because browsers can't call most MCP servers directly. Header values are sent to the server for that request only and are never stored or logged.
Tool list from our last check. Connect live to refresh it.
Check Arabic, Urdu or Persian text for problems that break PDFs, search and copy-paste: pre-shaped glyph characters and reversed letter order (typical of text copied out of a broken PDF), Persian or Urdu letters mixed into Arabic and the reverse, hidden direction-control characters, mixed digit systems, and characters the fonts cannot print. Returns the issues in English and Arabic and, when a safe fix exists, a fixed version of the text for the user to review. Deterministic rules; no AI.
The text to check.
Language of the text, or 'auto'.
Pick your client. Values in capitals, like YOUR_TOKEN, are yours to fill in.
Run in a terminal
claude mcp add --transport http arabic-pdf https://rtlpdf.com/mcp
Check Arabic, Urdu or Persian text for problems that break PDFs, search and copy-paste: pre-shaped glyph characters and reversed letter order (typical of text copied out of a broken PDF), Persian or Urdu letters mixed into Arabic and the reverse, hidden direction-control characters, mixed digit systems, and characters the fonts cannot print. Returns the issues in English and Arabic and, when a safe fix exists, a fixed version of the text for the user to review. Deterministic rules; no AI.
| Argument | Type | Description |
|---|---|---|
| text* | string | The text to check. |
| language | string | Language of the text, or 'auto'. |
{
"type": "object",
"title": "check_arabic_textArguments",
"required": [
"text"
],
"properties": {
"text": {
"type": "string",
"title": "Text",
"maxLength": 200000,
"minLength": 1,
"description": "The text to check."
},
"language": {
"enum": [
"auto",
"ar",
"ur",
"fa"
],
"type": "string",
"title": "Language",
"default": "auto",
"description": "Language of the text, or 'auto'."
}
}
}Create a PDF from Arabic, Urdu or Persian text, or text that mixes them with English, with every letter correctly joined, lines running right to left, and numbers, dates, phone numbers and Latin words kept in order. Suited to PDFs, printable files, letters, notices, invoices, certificates, menus and handouts in Arabic, Urdu or Persian. The text is printed exactly as given; light Markdown is supported: '# ' headings, blank lines between paragraphs, '- ' bullets, '1. ' numbered lists, '| a | b |' tables (a '|---|' row marks a header) and **bold**. Returns a download link that works for 24 hours. It only lays out text: it does not write, translate or change wording, and it does not make images.
| Argument | Type | Description |
|---|---|---|
| font | string | 'auto' picks Nastaliq for Urdu and Naskh otherwise. Cairo and Tajawal are modern sans styles. |
| text* | string | The full text to print, in Arabic, Urdu or Persian (English may be mixed in). Light Markdown allowed. |
| bleed | boolean | Add 3 mm bleed and crop marks for professional printing. |
| paper | string | Page size. |
| title | string | Document title for the PDF metadata and file name. Defaults to the first heading. |
| justify | boolean | Justify paragraph lines. |
| archival | boolean | Export as PDF/A-2a with PDF/UA-1 (archival and accessible), falling back to PDF/A-2b. |
| language | string | Language of the text. 'auto' detects Urdu and Persian letters. |
| font_size | any | Text size in points, 7 to 28. Leave empty for the font's default. |
| page_numbers | boolean | Print page numbers when there is more than one page. |
{
"type": "object",
"title": "make_pdf_arabicArguments",
"required": [
"text"
],
"properties": {
"font": {
"enum": [
"auto",
"naskh",
"nastaliq",
"cairo",
"tajawal"
],
"type": "string",
"title": "Font",
"default": "auto",
"description": "'auto' picks Nastaliq for Urdu and Naskh otherwise. Cairo and Tajawal are modern sans styles."
},
"text": {
"type": "string",
"title": "Text",
"maxLength": 30000,
"minLength": 1,
"description": "The full text to print, in Arabic, Urdu or Persian (English may be mixed in). Light Markdown allowed."
},
"bleed": {
"type": "boolean",
"title": "Bleed",
"default": false,
"description": "Add 3 mm bleed and crop marks for professional printing."
},
"paper": {
"enum": [
"A4",
"A5",
"Letter"
],
"type": "string",
"title": "Paper",
"default": "A4",
"description": "Page size."
},
"title": {
"type": "string",
"title": "Title",
"default": "",
"maxLength": 200,
"description": "Document title for the PDF metadata and file name. Defaults to the first heading."
},
"justify": {
"type": "boolean",
"title": "Justify",
"default": false,
"description": "Justify paragraph lines."
},
"archival": {
"type": "boolean",
"title": "Archival",
"default": false,
"description": "Export as PDF/A-2a with PDF/UA-1 (archival and accessible), falling back to PDF/A-2b."
},
"language": {
"enum": [
"auto",
"ar",
"ur",
"fa"
],
"type": "string",
"title": "Language",
"default": "auto",
"description": "Language of the text. 'auto' detects Urdu and Persian letters."
},
"font_size": {
"anyOf": [
{
"type": "number",
"maximum": 28,
"minimum": 7
},
{
"type": "null"
}
],
"title": "Font Size",
"default": null,
"description": "Text size in points, 7 to 28. Leave empty for the font's default."
},
"page_numbers": {
"type": "boolean",
"title": "Page Numbers",
"default": true,
"description": "Print page numbers when there is more than one page."
}
}
}We read every tool's name, description and inputs the way the model will, looking for instructions hidden from you. A clean result means no warning signs, not a guarantee. Scanned Nov 19, 2026.