pdf_add_page_numbers
Add page numbers to a PDF. Supports configurable position, format, starting number, and font size.
LABSCO SUMMARY
A PDF server that writes as well as reads — and can turn a scan into something a model can actually see.
Most PDF servers stop at extracting text, which leaves scanned documents useless. Rendering pages to images closes that gap by handing them to a vision-capable client instead. The other thing worth knowing is how carefully the page operations treat forms: merging, splitting and reordering keep AcroForm fields, collisions are namespaced by source, and each call reports what it preserved, renamed or dropped rather than losing them quietly. Encryption is AES-256 rather than legacy RC4, and none of it needs a build toolchain.
What it is
A write-capable PDF toolkit with 22 tools: extract text and metadata, convert to Markdown, search, merge, split, reorder and delete pages, fill and flatten forms, watermark, encrypt, and create documents from text, Markdown or a template. Every engine is WASM or plain JavaScript, so there is no native build step.
What you get
- pdf_extract_text returns text by page range, defaulting to the first ten pages so a long document does not swamp the context, and pdf_to_markdown reconstructs reading order across up to two columns with full-width title and footer bands handled separately
- pdf_render_pages rasterises pages to PNG or JPEG at a chosen DPI — the route to reading a scanned or image-only PDF with no text layer, using a vision-capable client
- pdf_search matches a literal substring across the document and returns the page number with a surrounding snippet; pdf_get_metadata returns title, author, subject, page count, dates and producer; pdf_compare diffs two PDFs page by page and reports where the text differs
- Page surgery that keeps forms intact — merge, split, reorder and delete all preserve AcroForm fields, colliding names are namespaced by source, and every call reports what it preserved, renamed or dropped
- pdf_get_form_fields lists fields with their types, current values and required status; pdf_fill_form handles text, checkbox, dropdown, radio and multi-select fields; pdf_flatten bakes the values into the page and removes the interactivity
- Creation three ways: pdf_create from plain text with wrapping and page overflow, pdf_create_from_markdown with headings, tables, lists and fenced code, and pdf_create_from_template for an invoice, report or letter from structured data validated against the template
- Presentation and marking: rotate pages, add page numbers with a chosen position and format, apply a diagonal text watermark with opacity and rotation, embed a PNG or JPEG at coordinates, and embed a QR code or barcode as qrcode, code128, datamatrix, ean13, pdf417 or azteccode
- pdf_encrypt applies AES-256 password protection through qpdf rather than the legacy RC4 scheme, with separate user and owner passwords
- Errors carry stable codes with stack traces kept internal, off-page placements are rejected instead of silently clipped, and oversized responses are truncated without breaking the JSON
Requirements
Node.js 20 or newer, launched over stdio as @aryanbv/pdf-toolkit-mcp with npx. No configuration file, API key, Docker or compiler, and it works offline — every engine is WASM or JavaScript, so there is no node-gyp step or prebuilt binary needed on Windows, macOS or Linux. Non-Latin text in a document you create or fill needs a fontPath pointing at a suitable font.
Setup effort
One command — npx -y @aryanbv/pdf-toolkit-mcp
USE CASES
01
Reading a document that has no text layer
Render the pages to images at a workable DPI and let a vision-capable client read them; text extraction handles the rest of the file.
02
Getting a paper into the conversation cleanly
Convert to reading-order Markdown so columns do not interleave, then search for the passage you actually wanted.
03
Assembling one document out of several
Merge the sources, put the pages in the right order and drop what does not belong — with the form fields surviving all three.
04
Filling in a form and sending it on
List the fields with their types, fill them, then flatten so the values are part of the page rather than editable.
05
Producing a document worth sending
Build it from Markdown or from a named template with structured data, then number the pages.
06
Sending something that should not circulate
Mark it with a diagonal watermark and apply AES-256 protection with separate user and owner passwords.
22 tools
Counted once with only startup setup. Other versions, modes or optional credentials may expose a different list.
pdf_add_watermark
Add a text watermark to PDF pages. Watermark is centered and rotated diagonally by default. Applies to all pages if no page range is specified.
pdf_compare
Compare two PDFs page by page (by absolute page index) and report text differences. Returns identical:true when text matches. Diffs content-stream-order text (not visual reading order), so it is best for same-layout documents; reflowed or multi-column PDFs produce noisy diffs. Inserting/deleting a page shifts all later pages and reports them as changed. Large diffs are trimmed (truncated:true) to fit the response limit.
pdf_create
Create a new PDF from text content with automatic line wrapping and page overflow. Supports A4, Letter, and Legal page sizes. Provide fontPath for non-Latin text (Arabic, CJK, etc.).
pdf_create_from_markdown
Create a rich, high-fidelity PDF from Markdown (CommonMark + GFM). Supports headings, bold/italic, links, ordered/bullet lists, tables, fenced code blocks, blockquotes, and horizontal rules.
pdf_create_from_template
Create a polished PDF from a named template (invoice, report, or letter). Pass structured data matching the template's fields; data is validated against the template's schema.
pdf_delete_pages
Delete specific pages from a PDF, keeping the rest in their original order. AcroForm fields on the remaining pages are preserved. Set flatten:true to bake field values into static content. Cannot delete every page.
pdf_embed_image
Embed a PNG or JPEG image into a specific page of a PDF. Supports custom positioning and optional scaling with aspect ratio preservation.
pdf_embed_qr_code
Embed a QR code or barcode into a specific page of a PDF at given coordinates. Supports qrcode, code128, datamatrix, ean13, pdf417, and azteccode.
pdf_encrypt
Encrypt a PDF with AES-256 password protection. Requires a user password to open. Owner password controls editing permissions (defaults to the user password).
pdf_extract_text
Extract text content from a PDF file. Returns first 10 pages by default to avoid exceeding LLM context limits. Use the 'pages' parameter for specific pages.
pdf_fill_form
Fill form fields in a PDF. Supports text, checkbox, dropdown, radio, and list-box (multi-select) fields. Provide fontPath for non-Latin text (Arabic, CJK, etc.).
pdf_flatten
Flatten a PDF's form fields, baking their current values into the page content and removing interactivity. Form-less PDFs are copied unchanged.
pdf_get_form_fields
List all form fields in a PDF with their names, types, current values, and required status. Returns hasForm: false for PDFs without forms.
pdf_get_metadata
Get metadata from a PDF file including title, author, subject, page count, creation/modification dates, and producer information.
pdf_merge
Merge multiple PDF files into one. AcroForm fields are preserved; fields whose names collide across inputs are auto-renamed (namespaced by source). Set flatten:true to bake field values into static content.
pdf_render_pages
Render PDF pages to images so a vision-capable client can read scanned or image-only PDFs. Writes PNG/JPEG files and returns their paths; set inline:true to return image blocks the model can see directly.
pdf_reorder_pages
Reorder pages in a PDF. Specify the new page order as a comma-separated string (e.g. '3,1,2'). Duplicates are allowed. AcroForm fields are preserved. Set flatten:true to bake field values into static content.
pdf_rotate_pages
Rotate pages in a PDF by 90, 180, or 270 degrees. Rotation is additive to any existing rotation. Rotates all pages if no page range is specified.
pdf_search
Search text across a PDF and return matches with a short surrounding snippet and the page number. The query is matched as a literal substring (case-insensitive by default). Searches all pages unless a range is given.
pdf_split
Extract specific pages from a PDF into a new file. AcroForm fields on the extracted pages are preserved; fields on omitted pages are dropped. Set flatten:true to bake field values into static content.
pdf_to_markdown
Convert a PDF to clean, reading-order Markdown for LLM consumption: reconstructs up to 2 content columns (plus full-width title/footer bands), infers headings from font size, and detects bullet/numbered lists. Pages with 3 or more columns fall back to single-column reading order. Tables are emitted as plain reading-order text, NOT reconstructed as Markdown tables. Best on clean, digital (text-based) PDFs; degrades on scanned/image-only PDFs (use pdf_render_pages for those) and very complex layouts. Returns the first 10 pages by default.
BEFORE YOU INSTALL
What to know first
Tools you get
22 tools when we connected. The five prompts, including pdf-to-markdown, and the pdf-toolkit://templates resource are not tools; the prompt pdf-to-markdown is separate from the tool pdf_to_markdown.
Requires
Node.js 20 or later. Node 18 and the 20.x line are end-of-life, so Node 22 or 24 LTS is recommended
Where it runs
All 5 documented clients run the server on your machine.
CONTEXT COST
Moderate Standing cost
Rated by the size of the 22 tool definitions that ride along on every turn, not by how many there are
INSTALLATION
Connect your client
Server setup
TERMINAL
npx -y @aryanbv/pdf-toolkit-mcpMCP Clients
Claude Desktop, Cursor, Windsurf
JSON
{
"mcpServers": {
"pdf-toolkit": {
"command": "npx",
"args": [
"-y",
"@aryanbv/pdf-toolkit-mcp"
]
}
}
}Claude Code
CLAUDE CODE · TERMINAL
claude mcp add pdf-toolkit -- npx -y @aryanbv/pdf-toolkit-mcpVS Code
VS CODE · JSON
{
"servers": {
"pdf-toolkit": {
"command": "npx",
"args": [
"-y",
"@aryanbv/pdf-toolkit-mcp"
]
}
}
}
