PDF & Documents
Local PDF Workflows: Merge, Extract, Organize, and Convert
Merge documents, extract selected page ranges, rotate or reorder pages, and turn JPG or PNG scans into a PDF locally in your browser without server uploads.
Document triage
Pick the page-tree change you need
Merge, extract ranges, reorder or rotate pages, or pack images cleanly without modifying source files.
Input
unencrypted PDFs or JPG/PNG images (≤25 MB)
Worker
local page-tree assembly via pdf-lib
Check
decrypted pages, orientation, and final copy
Everyday PDF paperwork rarely needs a desktop suite or a cloud upload. Merging separate reports, extracting selected chapters, rotating sideways scans, and packaging screenshots into one PDF all operate at the page-tree level inside an unencrypted document. Start from the PDF tools category or open the specific workspace — Merge PDF, Extract PDF Pages, Organize PDF Pages, or Images to PDF — so you copy only the bytes you need while your files are processed locally in your browser tab.
Choose the right PDF operation first
Picking the smallest operation that matches your task keeps output file sizes predictable and avoids re-exporting documents twice. Each Toolk PDF workspace handles a focused step on the underlying PDF page tree inside a background Web Worker powered by pdf-lib, with page thumbnails rendered locally via PDF.js.
| Task | Tool to use | Input rules | What changes in the new PDF |
|---|---|---|---|
| Combine multiple files into one packet | Merge PDF | 2 to 20 unencrypted PDFs, 25 MB total combined | Appends each document's pages in your chosen file order |
| Save specific pages or chapters | Extract PDF Pages | 1 unencrypted PDF, up to 200 pages and 25 MB | Copies selected page numbers or comma-separated ranges in order |
| Fix sideways scans, delete blanks, or reorder | Organize PDF Pages | 1 unencrypted PDF, up to 200 pages and 25 MB | Reorders the page array, drops removed pages, and adds 90° rotations |
| Convert receipts or screenshots into a PDF | Images to PDF | Up to 20 JPG or PNG images, 25 MB total, max 20 MP each | Places one image per page on A4, US Letter, or image-fitted pages |
Because every workspace writes a brand-new file instead of modifying the original in place, keep your source files archived until you have opened and checked the downloaded copy in a PDF viewer.
Workflow 1: Combine separate reports and appendices
Invoices, signed cover letters, and technical appendices often arrive as separate PDFs that must travel as a single attachment.
- Open Merge PDF and select your source documents (up to 20 files within the 25 MB combined limit).
- Reorder filenames using the earlier/later controls so the executive summary or cover sheet sits first.
- Run the merge inside the browser worker and inspect the preview thumbnails before downloading.
Each input document keeps its internal page order intact. If a merged file also needs an internal page rotated or a trailing blank page removed, pass the downloaded output through Organize PDF Pages as a quick second step.
Workflow 2: Extract a subset without exposing extra pages
Sharing a 120-page manual when a stakeholder only needs pages 14–19 wastes bandwidth and risks leaking unrelated sections.
Extract PDF Pages maps page selections to zero-based PDF page indices starting at file position 1:
- Ascending ranges (
2-5): copies pages 2, 3, 4, and 5 in sequential order. - Custom ordering (
9, 2-4): places page 9 first, followed by pages 2 through 4. - Repeated references (
1-3, 2, 5): deduplicates page 2 automatically after its first appearance so output pages do not repeat unintentionally. - Descending or out-of-range indices (
8-4or0): rejected upfront with a validation message before worker execution.
Remember that physical PDF file order (1, 2, 3...) can differ from printed Roman numerals (i, ii) or appendix labels (A-1). Clicking the preview thumbnails rendered by PDF.js helps verify exact pages when preface numbering shifts printed page numbers.
Workflow 3: Rotate sideways scans and trim clutter
Scanner feeders frequently produce mixed portrait and landscape pages or insert blank backsides. Organize PDF Pages lets you adjust the page array without altering selectable text on unaffected pages:
- Rotate 90° clockwise: adds 90 degrees onto each page's existing rotation dictionary entry (
0°,90°,180°, or270°). - Move earlier or later: shifts a page object earlier or later in the output document tree.
- Remove page: omits blank separator sheets or duplicate scans while ensuring at least one page remains.
- Restore original pages: discards unsaved workspace edits immediately if you decide to restart from the uploaded file order.
Workflow 4: Turn images and phone scans into a clean PDF
When receipts, whiteboard photos, or UI screenshots need to become a shareable document, Images to PDF packs JPG and PNG assets onto one-image-per-page layouts.
Before generating the PDF, consider two image preparation steps that save significant file weight:
- Shrink oversized camera photos first. High-resolution photos can eat into the 25 MB batch limit quickly. Run large PNGs or JPEGs through the image compressor first to keep text legible at a fraction of the byte size.
- Verify proportions for print framing. If you are cropping scanned receipts or banners to fit US Letter (
612 × 792 pt) or A4 (595.28 × 841.89 pt) cleanly, check proportions with the aspect ratio calculator.
Inside Images to PDF, A4 and US Letter add a 24-point margin and center each image proportionally without stretching, while Fit page to image computes exact page dimensions at 96 pixels per inch with zero margin. Because phone EXIF rotation metadata is not automatically rewritten during embedding, make sure photos already appear upright in your file browser before importing them.
Know the technical boundaries and privacy model
Understanding what client-side PDF object manipulation can and cannot preserve protects you from subtle document regressions:
- No password-protected PDFs or AcroForms: Encrypted documents cannot be parsed without decrypting first, and interactive fillable forms are rejected to avoid broken field bindings. Export a flattened or unencrypted copy from your OS print dialog first if needed.
- Document-level metadata changes: Creating a clean page-tree output does not preserve cryptographic digital signatures, interactive form scripts, outlines/bookmarks, or file attachments. Keep the original signed PDF whenever signature verification is legally required.
- OCR and word editing are out of scope: These tools manipulate page objects and embedded images; they do not perform optical character recognition or edit paragraph text inside existing PDFs.
- Local browser processing vs analytics: PDF bytes are read and rewritten inside a local Web Worker tab; Toolk's servers do not receive your file contents, and site page-view telemetry stays separate from workspace inputs.
Continue with related tools and guides
- Browse every document utility in the PDF tools hub or explore adjacent browser utilities in our developer tools workflow guide.
- Need to embed tiny icons or screenshots directly into HTML or CSS instead of a PDF packet? Read the Base64 encoding guide or pair it with Image to Base64.
Keep reading
Related posts
Words to Pages: Estimates for 500, 1000 and 2000 Words
Estimate pages from word count using explicit layout assumptions. Compare examples and learn when spacing, fonts, margins, tables, and images change the result.
Read postText Editing Workflow: Count, Clean and Format Content
Use text tools to count words, clean pasted content, compare revisions, and prepare Markdown. Preserve meaning and formatting through each editing step.
Read postPreparing Web Images, QR Codes, and Alt Text Audits in the Browser
Reduce image byte weight with HTMLCanvasElement, choose resilient QR code error correction, and audit missing alt attributes directly in the browser.
Read post