A personal PDF editor is mostly libraries: PyMuPDF for page changes, form fields, signature stamps and true redaction, OCRmyPDF for scans and PDF.js for viewing. An agent can wire that into a local app in a weekend. What stays rough is the thing people buy Foxit for, which is changing text in place. PDFs store positioned glyphs, often in subset fonts, so a short edit usually works and a reflowed paragraph does not. Word and Excel conversion that keeps layout and certificate-based signatures are the other gaps.
Before you buildFoxit has a 14-day trial, not a free plan. If you only merge, split or annotate, Stirling-PDF or PDFsam may already be enough.
Build me a local PDF editor to replace Foxit PDF Editor. Requirements: - Python 3.12 + FastAPI serving one HTML page with vanilla JS on http://127.0.0.1:5300. PDF_DIR in .env is the library: a sidebar lists its PDFs and opens one. Every save writes <name>-edited.pdf next to the original and never overwrites it. - Viewer: PDF.js (`pdfjs-dist`, vendored under /static) with zoom, a thumbnail strip and search. Edits are sent to the server, applied with PyMuPDF (`pymupdf`), and the page redraws from the new file. - Text edit: click a text span and type a replacement. The server removes the old span and inserts the new text in the same font and size when the font has the glyphs, otherwise in a bundled Liberation or Noto font, and shows a notice naming the fallback. No paragraph reflow. - Pages: delete, rotate, drag to reorder, insert pages from another PDF, extract a range, merge several files. - Forms and signing: list AcroForm fields (`page.widgets()`), fill them in the UI, flatten on request; place a PNG signature or a typed name as an image stamp on any page. - Redaction: draw boxes or search for a string, apply with `add_redact_annot` and `apply_redactions`, then re-read the page text to confirm the covered words are gone. - OCR for scans through `ocrmypdf` with a language picker, producing a searchable PDF and showing the size before and after. - Compare two PDFs: a page-by-page word diff of their text (`difflib`). - Single user. No accounts, no telemetry, local-first. - Out of scope: paragraph reflow, certificate-based digital signatures, Word or Excel export, mobile apps, cloud storage. - README: installing Tesseract and Ghostscript, where files go, PyMuPDF's AGPL-3.0 licence, and a warning that edits in subset fonts can fail or look different.
Opens your agent with the prompt filled in, you press enter. Or copy it and paste it into an empty folder.
The first prompt builds the core. Run these one at a time in the same folder to grow it into your own Foxit PDF Editor.
Extend the PDF editor in this folder. Requirements: - Add a Bulk redact page: pick a folder, tick patterns (email, phone, IBAN, a custom regex) and list every match with its file and page before anything changes. - After confirming, apply redactions to copies in ./redacted/ and re-read each result to check the matched text is gone; report any file where it is not. - Keep a CSV log of file, page and pattern, with no matched text in it. - Update the README.
Extend the PDF editor in this folder. Requirements: - Add a Mail merge page: choose a PDF with form fields and a CSV; map columns to field names with a dropdown per field, remembered per form. - Generate one filled PDF per row into ./merged/, named from a column you pick, flattened if ticked. - Show the first row as a preview before generating, and stop with a clear message if a field name is missing. - Update the README.
Extend the PDF editor in this folder. Requirements: - Add a Stamp tool for one file or a folder: Bates numbering with a prefix and starting number, a diagonal text watermark, and a header or footer with page x of y. - Position by preset (corners, centre) with a margin in millimetres, font and opacity. - Write results as new files and show the first page of the first result as a preview. - Update the README.
People who live in PDFs want one tool that opens whatever arrives, edits the text on messy real-world files, converts to Office and signs, without learning a new one. Foxit sells that Acrobat-style workflow at a lower price and adds volume licences and a perpetual option for IT departments.
Estimate · Runs on your machine with nothing to host. The build is token usage on your agent plan; expect more of it if you keep tuning the text editing.
| Plan | Monthly | Annual, per month | What you get |
|---|---|---|---|
| PDF Editor | $10.99/user | $10.83/user | Edit text, images and pages, convert to and from Office files, build and fill forms, OCR, compare two PDFs, password protection, Foxit DMS, Windows, macOS and web apps, 300 free AI credits a month. |
| PDF Editor+ | — | $13.33/user | Everything in PDF Editor plus iOS and Android apps, the cloud web app, AI Smart Redact for personal data and Foxit eSign with 150 envelopes a year. |
| PDF Editor 14 (perpetual) | custom | — | Windows or macOS only: editing, conversion, OCR, forms, security and fill-and-sign. No AI Assistant, Workspace, Smart Redact, DMS or cloud features. |
| AI Assistant add-on | — | $4.17/user | 2,000 AI credits a month for summaries and questions about a document. |
No free plan; every plan starts with a 14-day trial with no card needed.
Kinda. The core is a weekend build with the prompt on this page, but real gaps remain: Paragraph reflow and clean text editing in subset or unusual fonts, Layout-faithful PDF to Word, Excel and PowerPoint conversion.
About $10.99 a month on the PDF Editor (one-year plan, billed monthly) plan per seat, $131.88 a year (checked 2026-10-19).
Paragraph reflow and clean text editing in subset or unusual fonts; Layout-faithful PDF to Word, Excel and PowerPoint conversion; Certificate-based digital signatures and tracked eSign envelopes; The iOS and Android apps and cloud document management with check-in and check-out; AI summaries and Smart Redact for detecting personal data. If any of those matter to you, keep paying.
Yes: Stirling-PDF (open source), PDFsam Basic (open source), PDF Arranger (open source).