Privacy
Your documents
The online edition processes PDF contents, images, form values, OCR results, filenames, and edit history on your device. The application does not upload them to the project, an AI provider, or a document-processing server. The desktop edition sends PDF data to its own authenticated service on 127.0.0.1, on the same computer, to edit original text and images. It does not send that data to a remote service.
Open documents, recent files, stars, and history are held for the current application session. Save your work before closing or refreshing. The project does not receive your documents and cannot recover them. Saved downloads, printed copies, browser/OS caches, crash recovery, swap, backups, and files in folders synchronized by another service are controlled by your device and its software.
Browser storage
The application stores your light/dark preference in localStorage under pdfcabin-theme and your interface language under pdfcabin-language. The website and online app share the language preference on the same origin. Translation catalogs are bundled application assets, with no remote translation service; changing language does not translate or upload document content. The desktop app also stores the language in language.json inside its application user-data directory, independently of its local server port. Tesseract may cache the English OCR model in IndexedDB; this model cache is not your document or recognized text. Browser caching also stores application assets. Clear this site's storage in your browser settings to remove its local preferences and cached assets. The application does not add advertising cookies, analytics, tracking pixels, or fingerprinting.
Website hosting and external links
The public website and online app are hosted on Cloudflare Pages at pdf-cabin.pages.dev. A GitHub Pages copy remains available. Visiting either sends normal web requests to the respective host and its delivery infrastructure, including your IP address and browser/request metadata, for delivery and security. GitHub documents logging visitor IP addresses for security. Downloads and links to GitHub, documentation, or other sites are subject to those providers' policies. Hosting data is separate from local PDF processing. A copy hosted on ChatGPT Sites is subject to that host's access controls and privacy practices; this project does not control its infrastructure logs.
The project adds no account system, advertising, payments, mailing list, or remote crash-reporting service. Application setup and builds download software dependencies and an OCR model; document processing uses the copies served with the application.
- Cloudflare privacy policy
- GitHub privacy statement
- GitHub Pages data collection
- OpenAI privacy policy for the optional Sites host
Contact, choices, and updates
The project is maintained by Juan Solano / JSolanoDev. For a privacy or rights inquiry, use the project contact link below. GitHub issues are public: do not attach private PDFs, identity documents, financial data, credentials, or other sensitive information. Ask for a suitable private contact method before sharing personal details. Information you choose to post on GitHub is processed under GitHub's policies and can remain in its history; do not assume closing an issue erases it.
You can use the downloadable desktop app without visiting the online workspace, clear browser storage, and choose whether to follow external links. For requests about hosting logs or GitHub account information, contact the relevant provider. Applicable privacy rights remain available under your local law. This notice describes the current release; changes to data handling require an updated notice before the new behavior ships.
Terms & responsible use
Software license
PDF Cabin's original project code is offered under the MIT License. Third-party software keeps its own licenses and notices. These website notices do not add restrictions to, or replace rights granted by, those open-source licenses. No subscription, purchase, or account is required. Keep the applicable copyright and license notices when redistributing software.
Your files and permissions
You keep your rights in your documents. The project does not claim ownership of them or obtain a license to train on them. Only process documents you own or are authorized to use, and follow the copyright, privacy, confidentiality, and other rules that apply to your work. Do not use the hosted service to impersonate others, forge authorization, infringe rights, or attack the service. Editing a document does not grant permission to distribute its contents. Software-license rights remain governed by the licenses themselves.
Review results and keep originals
PDF formats and fonts vary. Editing, conversion, and OCR can change layout, lose information, or produce incorrect text. Keep an original, inspect the saved output, and verify names, amounts, dates, and other important details. Save before closing the app. The project does not promise compatibility with every PDF, uninterrupted hosting, or recovery of unsaved work.
Cropping, white overlays, and removing visible objects are not secure redaction or a guarantee of erasure. A drawn signature is a visual mark; the app does not verify identity or provide certificate-based digital signing. Editing may invalidate existing digital signatures. This app does not certify a document's authenticity, accessibility, legal validity, or regulatory compliance.
Warranty and legal rights
The software is supplied as is, subject to the warranty and liability terms in the applicable licenses. To the extent permitted by applicable law, the maintainers make no additional guarantee of accuracy, fitness for a particular purpose, or non-infringement. Nothing here excludes a right, remedy, warranty, or liability that the law does not allow to be excluded. These notices do not establish forced arbitration, a chosen court, or a waiver of mandatory consumer rights. Use of this application is not legal, financial, medical, or other professional advice.
Independent project and rights concerns
PDF Cabin is an independent project. It is not affiliated with, sponsored by, endorsed by, or certified by Adobe Inc. Adobe and Adobe Acrobat are trademarks of Adobe Inc. Other product names belong to their respective owners. References to another product do not grant permission to use its branding or proprietary assets.
For a copyright, trademark, privacy, or other rights concern, contact the maintainer with the relevant public URL, a description of the concern, and the rights you represent. Do not post private documents. This contact route is not a registered DMCA agent, a promise of legal safe-harbor status, or a legal determination of your claim.
AI & OCR transparency
AI-assisted development
AI coding assistance was used in developing PDF Cabin and its website. That describes the development process; it does not mean a person opening a PDF is chatting with an AI or sending a document to the coding assistant. AI assistance does not guarantee originality, security, correctness, or legal clearance. The project makes no claim to exclusive copyright in purely AI-generated expression. Contributors remain responsible for the rights, provenance, and review of what they submit.
What runs when you use the app
The current app has no generative-AI chat, document summarizer, cloud AI service, or automated decision-making feature about people. English OCR uses Tesseract's machine-learning text recognition locally on your device. The project does not collect your documents or OCR results to train models, and the app has no document-training pipeline. The English recognition model is distributed with its license and may be cached by your browser.
OCR is an estimate
OCR can omit, duplicate, or misread words, numbers, handwriting, tables, and unusual layouts. Confidence values are estimates, not guarantees. Compare recognized text to the source image before relying on or sharing it, especially for important documents. OCR currently exports text; it does not replace scanned text as native PDF objects or create a searchable PDF layer.
Future AI features
A future feature that sends document data to a provider must explain the provider, purpose, data sent, retention/training terms, and relevant choices before that transfer, and obtain consent when required. AI interactions and generated content must be identified where applicable. New AI features require a fresh privacy, licensing, and legal assessment before release. This notice is not a statement that every AI law applies to this release or that the app has been certified under any AI regulation.
Licenses & notices
PDF Cabin is MIT licensed. Third-party components retain their own terms. These notices do not certify legal clearance or regulatory compliance.
PDF Cabin is independent of Adobe Inc. Adobe and Adobe Acrobat are trademarks of Adobe Inc.