PDF tools for lawyers — contracts, exhibits, redaction and archival
Updated 15 August 2026
Legal documents are the ones you cannot hand to an unknown server: privileged correspondence, draft contracts, exhibit bundles. UnboundPDF redacts, paginates, compares and archives them inside your browser, so no copy leaves your device. Whether a court or client rule permits a given step is yours to check — this page is not legal advice.
Free PDF tools for legal work: redact properly, strip metadata, paginate exhibits, archive to PDF/A — processed in your browser, never uploaded.
UnboundPDF is a free suite of 43 PDF and image tools that run entirely in your browser — merge, split, compress, edit text, OCR in 126 languages, redact, sign, convert and archive to PDF/A. Your document is read and written by the page on your own device; there is no document-upload endpoint in the core tools, no account, no watermark and no daily cap. Every result can be checked — with the Network tab, or with the Document Passport the Workspace writes for a chain of steps.
- Open Redact PDF and add the document you are producing to the other side.
- Draw boxes over privileged or personal passages, keeping rasterise enabled so the text is removed.
- Clear the document properties in Edit Metadata, then merge the exhibits into the bundle.
- Add page numbers across the bundle, and archive the final version with PDF/A.
Client files are the category where "we delete uploads after an hour" is not a satisfying answer: privileged correspondence, unfiled drafts, third-party personal data inside an exhibit. Processing in the browser means no copy of the document is transmitted to us at all — an architectural fact you can verify, not a policy you have to trust. What your professional rules permit is yours to check; this page is not legal advice.
Redaction that survives being parsed
The recurring disaster is a black rectangle drawn over text that anyone can select, copy or extract with a script. Redact PDF defaults to rasterising the pages you mark, re-rendering them so the covered text no longer exists in the file rather than sitting behind a shape. Keep that box checked. Two costs come with it, and you should price them in: a rasterised page loses its selectable and searchable text and usually grows in size, and only the pages you actually marked are re-rendered — untouched pages pass through with their text intact. Never substitute Crop PDF, which hides content outside the box without deleting it, and read can deleted PDF text be recovered before you decide a cover-up is sufficient.
Metadata and markup leave their own trail
A production copy can disclose more in its properties than on its face: the author field naming a colleague, an internal matter reference in the title, stale keywords from a precedent. Edit Metadata loads the current title, author, subject, keywords and creator so you can clear them before the file goes out. Remove Annotations strips comments, highlights, sticky notes, links and form fields — the review markup on a draft is exactly what should not travel. Markup drawn into the page content itself cannot be stripped, and the tool tells you rather than implying a clean sweep. When markup should stay but stop being editable — a stamped approval, a client's own comments — flattening bakes it into the page instead of removing it.
Bundles, exhibits and pagination
Merge PDF assembles the bundle in the order you set, Organize PDF removes duplicates and blanks from a thumbnail grid, and Add Page Numbers stamps consistent numbering across the whole pack with a live preview of page one — so a hearing reference points at one page and not two. When a single exhibit must be produced separately, Split PDF extracts the range, and Split by Size keeps parts under a filing portal's cap. The privacy mechanics of assembling a pack this way are in merging PDFs privately.
Versions, scanned evidence and archival copies
To see what moved between two drafts, Compare PDF shows them side by side and highlights differences — a visual, pixel-based comparison, useful for spotting change, not a semantic clause diff. Scanned evidence needs a text layer before it can be searched: OCR PDF recognises it in your browser and lays an invisible layer over the original pages, with the OCR guide covering what quality to expect and text extraction covering the copy-out. For long-term storage, PDF/A adds archival metadata aimed at PDF/A-2b — a step toward an archive-ready file, which is how we describe it rather than promising certified conformance. Confidential copies can be locked with Protect PDF (real AES encryption, unrecoverable password) or marked with Watermark PDF.
Scope
The signature tools place a drawn or typed signature image on the page — not a PKCS #7 certificate signature; if your matter needs qualified or certified signing, use a product that provides it. Redact PDF removes the text from the page's content and verifies on the finished bytes that it is gone. PDF/A output is PDF/A-2b, checked against our published PDF/A-2b rule list on your device before it claims conformance; DeviceCMYK artwork is reported rather than converted. Large scanned bundles are processed on your own machine: a 300-page, 119 MB scan and a 500-page bundle both completed on 2026-08-15 across three browser engines, every output reopened in an independent PDF reader. In the Workspace, every step records the SHA-256 of its output in a Document Passport anyone can recompute.
Prove the no-upload behaviour before the first real file
Run a test document through with DevTools open and watch the Network tab: the page fetches its own code, and no request carries your document bytes. The PDF Privacy Lab sets that check out step by step, and it is the kind of evidence an IT or risk review can repeat independently. Neighbouring desks: PDF tools for HR shares the redaction and metadata workflow, while students and teachers cover education. Sending a bundle by email? Compress it first.
Frequently asked questions
Is a black box over text enough to redact a PDF?
No — not on its own. A drawn box leaves the words underneath extractable by anyone who selects or parses the page. Redact PDF's default rasterises the pages you marked, re-rendering them so the covered text no longer exists in the file. Only marked pages are rasterised.
What do I lose when a page is rasterised?
That page becomes an image: no selectable or searchable text on it, and usually a larger file. Untouched pages keep their original text and quality. It is a deliberate trade — the text being genuinely gone is the point.
Can I sign a contract with this?
You can place a drawn, typed or uploaded signature image on the page with Sign PDF or Fill & Sign. It is not a cryptographic PKCS #7 certificate signature and carries none of that tamper-evidence, so check what your counterparty, court or jurisdiction requires before relying on it.
How do I paginate an exhibit bundle?
Merge the exhibits in order, then run Add Page Numbers to stamp consistent numbering in the position and format you choose, with a live preview of page one. Split PDF pulls a range back out if an exhibit has to be produced on its own.