PDFit.pro
Dashboard

Verified releases

What changed, and what is actually live.

Every entry reflects a tested release or a documentation milestone. Planned work stays on the roadmap until it passes its acceptance criteria.

Beta Hardening · Sep 7, 2026

Safer document processing and cleaner search paths

  • Changed every onsite tool card to a direct PDFit.pro route while keeping privacy-safe first-party click measurement and campaign aliases.
  • Aligned visible breadcrumbs with structured data, removed invented sitemap dates, and kept runtime API responses out of search results.
  • Added atomic daily processing reservations, restart-safe job accounting, and correct success or release behavior across UTC midnight.
  • Separated temporary jobs, Workspace sessions, and durable documents so cleanup cannot cross storage boundaries.
  • Added browser-independent Workspace version finalization plus guarded deployment, paired backups, and a split-storage rollback floor.
  • Kept live checkout disabled until provider, legal, tax, and native-processing isolation gates are independently verified.
Bing Metadata Release

Search snippets explain every public page clearly

  • Expanded all 35 indexable page titles and descriptions with unique, page-specific language instead of repeated keyword filler.
  • Kept rendered titles between 45 and 65 characters and descriptions between 145 and 165 characters for useful search-result context.
  • Added build and release gates that reject short, oversized, or duplicated indexable metadata before deployment.
  • Preserved every canonical URL, noindex boundary, sitemap route, social card, and structured-data graph while refreshing sitemap modification dates.
Beta Clean Pass

All beta tools stay free; clean exports become the only paid extra

  • Added a small, idempotent PDFit.pro mark to generated PDF results across direct tools, queued jobs, Workspace, and protected documents.
  • Added one $4.99 payment that removes the mark from new PDF exports for 365 days without automatic renewal.
  • Kept processing, page, storage, document, and version limits unchanged—Clean Pass is not a hidden Pro plan.
  • Added a reusable clean_exports entitlement, duplicate-purchase protection, signed webhook reconciliation, and coordinated fail-closed activation.
  • Completed the new pricing and account experience in ten languages and corrected malformed absolute URLs in search structured data.
Search Intent Release

Twenty real tools become complete search destinations

  • Added a unique three-step guide, three practical use cases, two honest questions, and three direct next-step links to every released tool page.
  • Aligned homepage, catalog, and tool metadata with natural task searches while keeping every free, format, privacy, and limit claim verifiable.
  • Published matching SoftwareApplication, HowTo, and FAQPage facts without fabricated ratings, hidden copy, doorway pages, or mass-generated empty routes.
  • Built a 60-link internal workflow graph across real tools while keeping all 35 canonical URLs stable.
  • Added static and deployment gates that reject missing visible content, mismatched schema, indirect related links, or stale sitemap dates.
Stable Release

Workspace revision workflow reaches Stable

  • Graduated multi-layer editing, named checkpoints, visual comparison, restore with undo, threaded review comments, and verified export after production gates passed.
  • Promoted the focused upload-to-download tool catalog after every stable-release acceptance gate passed.
  • Verified all 18 production PDF operations, four friendly rejection paths, and all 173 backend integration tests.
  • Verified all 41 routes at desktop and mobile widths, 84 internal links, ten interface languages, and zero horizontal page overflow.
  • Kept OCR, PDF to Excel, Community, API, and connected account workflows clearly labeled as previews.
  • Confirmed healthy API and worker containers, automatic monitoring, daily smoke tests, database backups, bounded logs, and rollback-ready deployment.
Public Beta

PDFit is open for public testing

  • Opened the core upload-to-download PDF tools for real public use without requiring registration.
  • Added a site-wide beta notice, a three-step first-use path, and one clear status page for live and preview capabilities.
  • Connected product feedback to the protected NovaCore support queue without publishing an email address or accepting document attachments.
  • Kept Workspace, OCR, PDF to Excel, Community, API, and connected account workflows clearly marked as previews.
  • Verified the complete public route inventory on desktop and mobile, all ten interface languages, production health, processing workers, and the live tool smoke suite.
Preview 1.38

Scanned tables reach the OCR fallback reliably

  • Treated an empty Poppler bounding-box document as missing text instead of spreadsheet content.
  • Prevented XML layout tags from masquerading as one-column rows and bypassing scanned-document recognition.
  • Added a regression for image-only PDFs while preserving the existing digital-PDF extraction path.
Preview 1.37

Choose the tables Excel should contain

  • Added a protected worker-backed preview for digital PDFs and scanned documents before workbook export.
  • Added bounded table samples, select-all and clear-all controls, validated table IDs, and export of only the chosen regions.
  • Used real PDF bounding boxes for column geometry, removed exact repeated headers, and merged nearby shared headings across table columns.
  • Stored safe numbers, currencies, negatives, and percentages as usable Excel values while preserving leading-zero identifiers and dates as text.
Preview 1.36

Side-by-side tables become editable blocks

  • Detected two unrelated tables separated by a clear, repeated horizontal gutter on the same PDF page.
  • Normalized the detected tables into top-to-bottom worksheet blocks in left-to-right reading order, with one blank separator row.
  • Preserved shared headings once, retained sparse rows and empty middle cells, and avoided splitting ordinary tables with uniformly wide columns.
  • Verified a real side-by-side PDF through Poppler, the XLSX package contract, and an independent office reader.
Preview 1.35

Separate tables keep their own columns

  • Detected vertically separated table regions on the same PDF page instead of applying one unrelated column layout to every row.
  • Kept one clear blank worksheet row between regions while preserving the one-page-per-sheet contract.
  • Counted region separators inside the existing workbook safety limits and retained text-safe identifiers and leading zeroes.
  • Verified incompatible three-column and two-column layouts plus a real two-table PDF through an independent XLSX reader.
Preview 1.34

File selection you can control

  • Added drag-and-drop file selection across every focused PDF tool and Workspace upload.
  • Added an explicit ordered file list for multi-file workflows such as Merge PDF and Images to PDF.
  • Added accessible move-up, move-down, and remove controls that update the actual upload order before processing.
  • Completed all new file-selection copy in ten languages and kept the compact layout free of horizontal overflow at 390 px.
  • Verified the production release in a real browser with two images, live reordering, removal, NovaCore localization, healthy API and worker containers, and clean browser logs.
Preview 1.33

An account that guides the next document

  • Rebuilt Dashboard with real document, version, storage, operation, plan, and account data instead of preview-only counters.
  • Added a four-step onboarding path that reflects actual email confirmation, completed tools, and saved Workspace documents.
  • Added document-name search, Active and Trash counts, current-version downloads, per-version sizes, guarded deletion, and clearer recovery states.
  • Completed email confirmation with resend from Settings, a safe one-use confirmation page, and exact allowlisted server routes.
  • Expanded the shared ten-language catalog and verified onboarding, storage, search, Trash, confirmation, and mobile layout in the browser.
Preview 1.32

Results that explain what happened

  • Standardized every focused quick-tool result with original and output sizes, page or file count, and a clear size change.
  • Added download again, process another file, and open editor actions without forcing users into Workspace for ordinary conversions.
  • Added safe summaries of selected processing options while excluding file content, passwords, and private document data.
  • Completed ten-language result vocabulary and a one-column 390 px mobile layout with equal-width actions and no horizontal overflow.
  • Verified the release against all 18 production tools, four friendly error paths, healthy API and worker containers, and clean browser logs.
Preview 1.31

OCR that understands page orientation and layout

  • Added automatic orientation and mixed-layout recognition for scanned documents.
  • Added focused standard-document, single-column, and sparse form or receipt modes when automatic recognition needs a hint.
  • Bounded every mode to a fixed server-side allowlist and packaged orientation data inside the isolated PDF worker.
  • Connected the layout choice to the scan prompt, Workspace settings, Job API, usage metadata, ten-language interface, tests, and deployment health gate.
Preview 1.30

Cleaner scans and faster quick tools

  • Added automatic deskew and gentle background cleanup before local printed-document recognition.
  • Added Balanced cleanup, Preserve appearance, and High contrast profiles to the scan prompt, Workspace settings, Job API, and Workspace step API.
  • Preserves the exact rendered page dimensions while keeping cleanup and OCR fully inside the isolated PDFit worker.
  • Added fixed profile validation, process and pixel bounds, ten-language translations, a user guide, and a deliberately skewed image-only quality fixture.
  • Extended the production deployment gate to verify both Tesseract and ImageMagick before a worker release can become active.
  • Standardized quick-tool results with verified PDF or ZIP downloads, a persistent result card, download again, process another file, and a direct Workspace continuation.
  • Restored the interactive page preview module across Organize, Rotate, Extract, Delete, and Reorder.
Preview 1.29

Searchable scanned documents

  • Added local printed-document OCR for English, Spanish, German, French, Italian, Portuguese, Polish, Russian, Ukrainian, and Turkish without sending document content to an external provider.
  • Workspace now detects likely image-only scans and offers a compact language-aware Recognize text action without blocking the rest of the editor.
  • Added OCR to the primary toolbar, categorized tool library, Workspace step API, Job API, capability discovery, and isolated worker dispatch.
  • Returns a searchable PDF directly to Workspace so recognized words can be selected, searched, copied, and handed to the existing Replace text workflow.
  • Bounded recognition by page count, rendered pixels, language allowlist, timeout, output size, output signature, and output page count while keeping OCR content and diagnostics out of analytics.
Preview 1.28

Guided editor experience

  • Reorganized the oversized editor toolbar into permanent essential actions and a categorized More tools library for create, review, forms, page, and finishing workflows.
  • Added an interactive Workspace tour that highlights the real controls in place, adapts to whether a PDF is loaded, supports Back, Next, Finish, and Escape, and remains mobile-safe.
  • Added direct forward and backward controls to every unlocked layer so touch users can change stacking order without relying on desktop drag and drop.
  • Improved advanced-tool active state, automatic menu closing, outside-click and Escape behavior, responsive bottom-sheet presentation, and source-contract regression coverage.
  • Separated visual Whiteout from Secure redact: whiteout now exports as a transparent full-page composition layer, while secure redaction visibly remains the irreversible flattened-data workflow.
  • Normalized whiteout brush geometry so its exported size follows the PDF page instead of changing with browser zoom, viewport size, or a later page resize.
  • Replaced the browser confirmation with an accessible export preflight that summarizes visible layers, affected pages, hidden objects, permanent edits, and visual-whiteout risk before processing.
  • Added post-export PDF verification for signature, page count, affected-page geometry, and file size, with an explicit report beside the editor.
  • Moved rulers behind a persistent View control, keeps them disabled on small screens, and lets users independently disable magnetic snapping without covering the PDF.
  • Connected the existing Fit, layer-order, export-preflight, and first-use tour systems into one cleaner desktop and mobile editing flow.
  • Replaced sequential editor export with one server-side atomic composition: every visible text, image, whiteout, replacement, link, comment, and form layer is committed as one document version only after the complete stack succeeds.
  • Kept a narrowly scoped compatibility fallback for older API deployments while production uses the single-step composition endpoint and leaves the current version unchanged on any failed layer.
  • Upgraded direct PDF text editing from fragmented glyph runs to complete selectable lines, with double-click replacement and document-search compatibility.
  • Automatically carries the detected font family, size, weight, italic style, line width, text color, and page background into the replacement layer.
  • Reduced the minimum text replacement height from five percent to one percent of the page so ordinary lines no longer become oversized cover boxes.
Preview 1.27

Workspace learning center

  • Added an in-editor quick guide that explains the complete open, choose, edit, review, apply, and download workflow without sending users away from their document.
  • Added a permanent Help control, first-use guidance, a tool chooser, essential keyboard shortcuts, and a mobile-safe dialog that can be reopened at any time.
  • Published a full Workspace guide covering editing, annotations, pages, forms, layers, stacking order, versions, precision alignment, manual guides, mobile use, secure redaction, export, and troubleshooting.
  • Connected the guide from the Workspace introduction and editor toolbar while preserving the focused editor, browser-local recovery, and no-account opening flow.
Preview 1.26

Precision alignment polish

  • Rebuilt multi-object alignment from visual page geometry so left, center, right, top, middle, and bottom commands align actual object edges instead of approximate stored coordinates.
  • Changed distribution into true equal-gap spacing while keeping the outermost selected objects anchored, and added Match width and Match height actions based on the primary selection.
  • Added keyboard rotation with bracket keys, 15-degree Shift steps, zero-angle reset, and double-click reset on individual and group rotation handles.
  • Improved transform status text with the live angle, compact shortcut guidance, clearer action grouping, disabled-state rules, and mobile-safe alignment menus.
  • Preserved locking, stacking, Undo/Redo, versions, local recovery, image proportions, and the exact worker-side PDF export contract.
Preview 1.25

True object rotation

  • Added direct rotation handles for text, images, drawings, signatures, shapes, stamps, and visual annotations, plus exact numeric angle input from -180 to 180 degrees.
  • Holding Shift snaps rotation to familiar 15-degree increments while the transform HUD reports the live angle next to X, Y, width, and height.
  • Added multi-object rotation around a shared selection center while preserving each object’s own angle, page geometry, stacking order, locks, Undo/Redo, named versions, and local recovery.
  • Carried object angles through the bounded API and worker pipeline so text and image rotations are written into the downloaded PDF instead of existing only in the browser preview.
  • Added PDF input validation, worker coverage, static editor contracts, and production-build verification for rotated exports.
Preview 1.24

Rulers and manual guides

  • Added compact rulers along the top and left edges of the active PDF page without reducing the usable document canvas.
  • Users can drag out vertical or horizontal page guides, reposition them by pointer or keyboard, remove one with Delete or a double click, and clear the current page from the ruler corner.
  • Manual guides join page edges, centers, and object geometry as first-class magnetic alignment targets for individual objects and selected groups.
  • Guides are page-specific and persist in the private seven-day browser-local recovery draft, but are never uploaded, analyzed, logged, or written into the exported PDF.
  • Added bounded guide limits, mobile-safe overflow behavior, keyboard access, production-build checks, and source-contract regression coverage.
Preview 1.23

Precision transforms

  • Added a live transform HUD that shows exact X, Y, width, and height percentages while moving or resizing one object or an entire selected group.
  • Extended magnetic alignment from movement to text and area resizing, including page edges, page centers, and neighboring object edges and centers.
  • Added professional modifier behavior: hold Alt while dragging to temporarily bypass magnetic alignment, or hold Shift while resizing text and free-form areas to preserve their proportions.
  • Keeps the geometry HUD inside the visible page, clears temporary guides after every transform, and preserves the existing undo, autosave, stacking, locking, and multi-selection model.
  • Added production-build and source-contract coverage for precision transforms without increasing the localization surface.
Preview 1.22

Desktop-grade canvas selection

  • Added drag-to-select directly on the PDF canvas: left-to-right selects fully enclosed objects while right-to-left also captures intersecting layers.
  • Preserves additive Ctrl, Command, and Shift selection and connects the result to the existing group move, resize, alignment, locking, visibility, ordering, duplication, and deletion tools.
  • Added an accessible right-click object menu for one layer or a selected group, including copy, duplicate, front/back ordering, lock, hide, and protected deletion.
  • Uses a compact anchored menu on desktop and a safe bottom action sheet on narrow mobile screens without covering the editor toolbar.
  • Expanded the NovaCore-owned ten-language catalog to 745 complete messages and added production-build and source-contract regression coverage.
Preview 1.21

Named versions and visual comparison

  • Added private named editor checkpoints that are stored with the existing browser-local recovery draft.
  • Compares the current composition with any checkpoint using green added layers, amber changes, red removed-object ghosts, and per-page totals.
  • Restores a checkpoint without losing the current edit: Undo returns directly to the composition from before the restore.
  • Keeps local editor checkpoints separate from durable server document versions and expanded the NovaCore-owned ten-language catalog to 743 complete messages.
Preview 1.20

Document review threads

  • Turned native review-comment layers into compact document discussion threads with replies, page-aware jump actions, and Open, All, and Resolved filters.
  • Added Resolve and Reopen states without flattening the editable comment layer or losing its author, geometry, stacking, undo, or browser-local recovery behavior.
  • Exports the bounded discussion transcript inside the native PDF comment so the review context remains available in compatible PDF readers.
  • Keeps comment and reply content out of analytics and public storage while preserving it in the existing private local draft until composition is applied.
  • Expanded the NovaCore-owned ten-language catalog to 710 complete messages and added browser coverage for reply, resolve, filter, reopen, responsive layout, and thread persistence.
Preview 1.19

Document-wide editor search

  • Added Ctrl/Command+F search across every text-bearing page without sending document text to the server or analytics.
  • Added previous/next navigation, active-page switching, visible match highlighting, case-sensitive search, and whole-word matching.
  • Connected every selected result directly to Replace and secure Erase so users can move from discovery to editing without manually drawing an area.
  • Kept the search strip compact and responsive, with no horizontal overflow at 390 px and an accessible keyboard-first interaction model.
  • Expanded the NovaCore-owned ten-language catalog to 689 complete messages and added four-page browser coverage for navigation, highlighting, case matching, replacement, localization, and mobile layout.
Preview 1.18

Safe form flattening

  • Added a dedicated Flatten mode that paints the selected appearance of every form widget into normal page content and removes the complete interactive AcroForm structure.
  • Preserves the original vector page, text, images, links, comments, and other non-form annotations instead of rasterizing the entire document.
  • Refuses to flatten completed certificate signatures because rewriting the PDF would invalidate their cryptographic integrity.
  • Validates every export by reopening it and requiring the original page count, no AcroForm dictionary, and no remaining Widget annotations.
  • Expanded the NovaCore-owned ten-language catalog to 671 complete messages and added engine, browser, payload, responsive, and structural PDF coverage.
Preview 1.17

Imported radio groups

  • Added automatic discovery of imported AcroForm radio groups, including parent fields whose actual radio widgets carry the export values.
  • Added a compact accessible choice control that preserves the current selection, exposes every available option, and works without horizontal overflow on desktop or mobile.
  • Updated the PDF worker to validate the requested export value, set the canonical group value, select the matching widget appearance, and turn every other widget off.
  • Explicitly rejects executable push-button actions instead of treating them as checkbox values, while preserving existing appearance streams and interactive export.
  • Added real two-widget radio-form coverage across the PDF engine, browser discovery, serialized workspace payload, production build, and responsive UI.
Preview 1.16

Discover and fill existing PDF forms

  • Added automatic PDF.js discovery of native text, checkbox, dropdown, read-only, and signature fields already stored inside an opened PDF.
  • Added a compact, responsive Forms panel that preserves current values, page references, multiline controls, dropdown options, and explicit certificate requirements.
  • Added a bounded fill-form worker operation that updates canonical AcroForm values, synchronizes widget appearance states, and rebuilds non-empty text, choice, and checkbox appearance streams.
  • Kept exported fields interactive instead of flattening them into pixels, while rejecting invalid choices, oversized payloads, and certificate signatures without a certificate.
  • Expanded the NovaCore-owned ten-language catalog to 662 complete messages and added backend AcroForm, browser discovery, payload, desktop, mobile, and production-build coverage.
Preview 1.15

Native fillable PDF forms

  • Added a dedicated Forms mode for placing real text inputs, checkboxes, dropdowns, and digital-signature fields directly on the Workspace canvas.
  • Kept every field movable, resizable, reorderable, copyable, lockable, hideable, undoable, and recoverable through the existing private local-draft layer model.
  • Exported canonical AcroForm fields and page Widget annotations with stable names, accessible labels, required/read-only/multiline flags, dropdown options, default values, and non-empty appearance streams.
  • Protected documents from ambiguous form trees with bounded values, strict field-name validation, duplicate-name rejection, page and geometry limits, and preservation of fields created in earlier steps.
  • Expanded the NovaCore-owned ten-language catalog to 645 complete messages and added structural AcroForm, rendering, responsive UI, localization, API, and production smoke coverage.
Preview 1.14

Native PDF review comments

  • Added movable review-comment layers with bounded author and comment fields, compact yellow note previews, ordering, locking, visibility, duplication, undo, and local draft recovery.
  • Exported every review note as a native PDF text annotation instead of flattening it into page pixels, so compatible PDF readers can open the original author and message.
  • Kept comment contents out of analytics, public responses, URLs, and application logs while preserving normal Workspace token, retention, and document-version boundaries.
  • Added strict page, geometry, text-length, and author validation plus backend inspection of the emitted PDF annotation dictionary.
  • Expanded the NovaCore-owned ten-language catalog to 618 complete messages and added desktop, mobile, localization, API, and production smoke coverage.
Preview 1.13

Links and document structure

  • Added real clickable HTTP and HTTPS link annotations that can be placed, resized, reordered, copied, locked, hidden, and exported from the visual Workspace.
  • Added editable PDF title, author, subject, and keywords metadata without exposing those values to product analytics or logs.
  • Added repeating headers and footers with page, page-count, title, and date tokens plus bounded top-level document bookmarks.
  • Kept link destinations protected by strict URL validation and rejected credentials, local protocols, invalid geometry, and out-of-range bookmark pages.
  • Expanded the NovaCore-owned ten-language catalog to 611 complete messages and added backend, browser, responsive, and production smoke coverage.
Preview 1.12

Fill & Sign inside the real editor

  • Added form-style text, date, checkbox, and initials fields that can be moved, resized, duplicated, copied, locked, hidden, reordered, and undone like every other Workspace layer.
  • Added mouse, pen, and touch signature drawing with transparent output, automatic trimming, and conversion into a movable image layer before export.
  • Kept the signature claim honest: this release provides visual signing and flattened PDF output, not certificate-backed identity verification.
  • Reused the existing isolated text and image worker pipeline instead of introducing a second export path, preserving bounded jobs, private temporary files, timeline history, and reliable downloads.
  • Expanded the NovaCore-owned ten-language catalog to 590 complete messages and added browser regression coverage for Fill & Sign, signature input, themes, desktop layout, and mobile overflow.
Preview 1.11

Annotations and complete page construction

  • Added editable highlight, underline, strikeout, and note layers with color, opacity, movement, resizing, ordering, visibility, locking, undo, autosave, and final PDF export.
  • Added visual crop framing with page selection and safe normalized geometry handled by the isolated PDF worker.
  • Added bounded blank-page insertion plus all-or-selected page insertion from a second validated PDF.
  • Kept the expanded editor compact on desktop and horizontally navigable on mobile, with the page strip below the full-width document canvas.
  • Expanded the NovaCore-owned ten-language catalog to cover the new document tools and added backend, browser, responsive, and localization release gates.
Preview 1.10

A real composition layer stack

  • Rebuilt the Workspace layer manager as a compact disclosure panel that keeps the document canvas visually dominant.
  • Made the list match the actual PDF stack, with the topmost object first and page, position, visibility, and lock state visible on every row.
  • Added one-step ordering plus explicit Send to back and Bring to front actions for single or grouped selections.
  • Preserved relative group order, composition undo/redo, local autosave, locked-object protection, and final PDF application order.
  • Added keyboard layer traversal, deletion, step ordering, boundary ordering, and responsive browser regression coverage.
Preview 1.9

One live capability contract and a sharper Workspace

  • Expanded the shared capability registry with formats, product surfaces, editions, security handling, API paths, and release status.
  • Changed Job API validation and plan upload limits to consume the shared contract instead of separate host lists.
  • Published cacheable capability discovery and a live developer table for clients, Nova deployments, and Standalone.
  • Fixed the missing production route for Documents to PDF and added release gates that reject the same mismatch in future packages.
  • Added Fit and 50–200% page zoom, four-corner object resizing, direct JPG/PNG drop onto the PDF canvas, and V/T/I/E editor shortcuts.
  • Stopped editing tools from automatically changing the page layout; Focus mode is now an explicit user choice.
  • Added real rectangle, ellipse, line, arrow, freehand drawing, and document-stamp layers using the existing isolated image-overlay worker.
  • Added one-click left, center, right, top, middle, and bottom alignment while preserving composition undo, layer order, and safe PDF output.
  • Added an object inspector with exact X, Y, width, and height controls, custom layer names, visibility, locking, and visible layer-state indicators.
  • Added internal object copy and paste across document pages with Ctrl/Command+C and Ctrl/Command+V while preserving private in-memory files and composition history.
  • Added seven-day browser-local autosave and crash recovery for unfinished object layers without storing the source PDF, plus explicit keep/discard controls and automatic cleanup after successful application.
  • Added PDF.js-powered selection of existing text and image regions: click content to replace it or erase it through the permanent redaction-first worker path, with a clean white replacement surface and replacement text sized from the detected source line.
Preview 1.8

Storage safety and independent Standalone foundation

  • Added visible plan-aware document, version, and byte quotas without counting an interrupted write as stored work.
  • Enforced document ownership inside each workspace so one member cannot enumerate or download another member’s private document.
  • Added daily orphan-artifact reconciliation with a safety window plus automatic removal of expired Trash artifacts.
  • Released a runnable NovaCore-free Standalone host with local-only defaults, guarded remote access, persistent documents, versions, audit, quotas, and maintenance.
  • Verified 114 backend tests, all ten localization checks, the production static package, both healthy production containers, exact document proxies, backups, and rollback gates.
Preview 1.7

Saved Workspace documents and recoverable versions

  • Added a private document library for signed-in Workspace sessions while keeping anonymous processing temporary.
  • Saved the original PDF and every completed Workspace action as an immutable, downloadable version with duplicate-safe step linkage.
  • Added version restore, Active and Trash views, 30-day recovery, automatic expired-trash cleanup, and tenant-scoped lifecycle audit.
  • Streamed document downloads without buffering the complete file in API memory and kept storage references out of public responses.
  • Verified 107 backend tests, all ten localization checks, the optimized 37-route web build, and a zero-warning solution build.
Preview 1.6

Visual workspace and secure redaction

  • Released a full-page workspace editor with desktop, mobile, and fullscreen modes instead of relying only on page thumbnails.
  • Added draggable text and JPG/PNG objects with page selection, pointer/touch/keyboard movement, and live size and position controls.
  • Added secure rectangular redaction that flattens the document before rebuilding it, so covered text, images, forms, annotations, and links are not retained in the downloaded PDF.
  • Kept redaction inside the bounded asynchronous worker with token-protected results, temporary render cleanup, plan limits, read-only containers, health gates, backups, and rollback.
  • Expanded the NovaCore-owned ten-language catalog to 456 complete messages and verified the live result through the worker down to a black RGB 0,0,0 pixel inside the selected area.
Preview 1.5

Private contact forms powered by NovaCore

  • Removed public support email addresses and mailto links from Support, Privacy, Terms, and the public Changelog.
  • Added a protected, attachment-free contact form for support, privacy, legal, security, billing, partnership, feedback, and other requests.
  • Stored requests in the tenant-scoped NovaCore queue with account linking, bounded fields, same-origin enforcement, per-IP rate limiting, anti-spam checks, and a silent honeypot.
  • Added a permission-gated NovaCore administrator inbox with new, in-progress, resolved, and spam workflow states.
  • Kept message text and reply addresses out of product analytics, link tracking, URLs, and public HTML.
Preview 1.4

Light and dark design everywhere

  • Added one compact system, light, and dark theme control to every public, tool, account, Community, legal, and error page.
  • Made device theme the default, saved anonymous choice between pages and reloads, and synchronized authenticated presentation with the existing NovaCore profile preference.
  • Expanded semantic theme tokens across navigation, cards, forms, tool results, status states, account surfaces, and Community surfaces.
  • Added localized accessible state, keyboard focus, mobile overflow coverage, early same-origin theme bootstrap, and release gates on all 30 pages.
  • Kept theme preference out of analytics and introduced no third-party request or new database field.
Preview 1.3

Reviewed translations for live workflows

  • Expanded the NovaCore-owned catalog from 280 to 380 complete ten-language messages for the homepage, tool catalog, and every live PDF workflow.
  • Rewrote awkward Russian product language and reviewed security, retention, format, option, and result guidance in context.
  • Added source-coverage checks so untranslated live-tool interface copy cannot silently pass a release, including a localized accessible file picker instead of browser-language buttons.
  • Kept brand names, file-format names, numbers, and language endonyms intentionally unchanged.
Preview 1.2

NovaCore-owned localization

  • Moved the initial 280-message, ten-language PDFit product catalog into a versioned NovaCore embedded resource with stable pdfit.ui keys.
  • Added canonical English source messages to the Core response so the static shell can consume tenant-aware translated values without duplicating runtime ownership.
  • Changed the browser runtime to validate and use the same-origin Core pack first, with a version-matched bundled snapshot only for Standalone operation or a Core outage.
  • Exposed one exact read-only localization route while stripping browser cookies and keeping every other non-allowlisted NovaCore route closed.
  • Added Core completeness tests, forced-outage browser coverage, catalog drift gates, and production checks for source, version, fallback, methods, headers, health, and rollback.
Preview 1.1

Ten-language interface

  • Released English, Spanish, German, French, Italian, Portuguese, Russian, Ukrainian, Polish, and Turkish interface support.
  • Translated the shared shell, catalog, live PDF-tool controls, primary account controls, and selected public-page introductions with an English fallback for detailed preview copy.
  • Added an accessible responsive language selector, URL propagation, browser-language detection, and a non-identifying local preference that connects to the saved NovaCore account language.
  • Kept every dictionary self-hosted without a third-party translation request and kept the language value out of product analytics.
  • Added ten-locale browser checks, mobile overflow verification, legal-language disclosure, and strict localization release gates on all 30 pages.
Preview 1.0

Search-ready public release

  • Verified real inbound delivery to the product support mailbox from an external Gmail account without a bounce or spam placement.
  • Published one exact canonical URL and complete Open Graph and large Twitter card metadata for every real page.
  • Published a 24-URL sitemap and crawler policy that exposes public content while keeping API endpoints out of crawl paths.
  • Kept Login, Registration, Dashboard, History, Settings, and not-found responses explicitly noindex.
  • Added release gates that reject missing canonicals, social metadata, sitemap drift, accidental account indexing, or a restored global Nginx noindex header.
Preview 0.9

Moderated Community

  • Released a public Community feed for focused ideas and reproducible bug reports, with category, type, status, newest, and popularity views.
  • Added optional account-bound publishing, idempotent upvotes, abuse reports, and a moderator workspace for post and case status changes.
  • Kept every query tenant-scoped through NovaCore and excluded post text, report details, IP addresses, user agents, and document data from product analytics and audit metadata.
  • Added same-origin mutation checks, exact JSON contracts, 8 KB request limits, bounded text and link rules, duplicate prevention, role gates, and four independent rate-limit policies.
  • Added idempotent production schema installation, rollback backups, self-cleaning end-to-end smoke coverage, and a strict Community release audit.
Preview 0.8

Privacy-safe product analytics

  • Added bounded first-party events for tool opens, tool-card clicks, job intent, and confirmed browser downloads.
  • Restricted ingestion to four events, fourteen tool identifiers, event-specific placements, same-origin requests, 2 KB bodies, and 30 requests per minute.
  • Excluded URLs, referrers, filenames, document data, credentials, IP addresses, user agents, and persistent visitor identifiers from the browser event contract.
  • Added one-time event deduplication, Global Privacy Control and Do Not Track support, exact Nginx allowlisting, and a self-cleaning production smoke test.
  • Kept existing server-side completion, rejection, failure, and Job API events as the authoritative processing outcome stream.
Preview 0.7

NovaCore account foundation

  • Released optional PDFit registration and secure login through a tenant-isolated NovaCore session.
  • Released responsive Dashboard, operation History, and Settings pages with explicit loading, signed-out, empty, success, and service-error states.
  • Added safe history that displays operation type, time, and status without document names, contents, passwords, access tokens, or watermark text.
  • Added saved light, dark, or system theme and ten-language preference support.
  • Kept all other NovaCore routes closed behind the exact Nginx account allowlist and added a self-cleaning production account smoke test.
Preview 0.6

Watermarks and page numbering

  • Released printable Unicode text watermarks with controlled opacity and orientation.
  • Released page numbering with six positions, three formats, and a custom starting number.
  • Added both operations to the protected persistent Job API without echoing watermark text.
  • Added a checked cross-platform font, page and text limits, safe analytics, and automatic cleanup.
  • Expanded the public catalog, deployment gates, and owned 4UL.ink tracking aliases.
Preview 0.5

Image and PDF conversion

  • Released ordered JPG/PNG to PDF conversion with automatic, A4, and Letter layouts.
  • Released 144-DPI PDF page export to numbered PNG or JPG files in one ZIP archive.
  • Added both conversion operations to the protected persistent Job API.
  • Added signature, decoded-pixel, render-pixel, output-size, rate, timeout, and cleanup controls.
  • Expanded owned 4UL.ink tracking aliases to every dedicated live tool.
Preview 0.4

Password protection and authorized unlock

  • Released Protect PDF with AES-256 encryption and encrypted metadata.
  • Released Unlock PDF for documents whose valid password the user supplies.
  • Kept passwords out of persistent jobs, URLs, logs, and NovaCore analytics.
  • Added safe password, file, page, rate, concurrency, timeout, and cleanup controls.
  • Added API, service, security, UI, deployment, and rollback verification.
Preview 0.3

Public product foundation

  • Introduced the shared website shell and canonical tool catalog.
  • Published product, roadmap, API, support, and legal preview pages.
  • Kept live tool entry cards routed through owned 4UL.ink tracking aliases.
  • Added a verified SVG favicon and release gates that protect it.
  • Established Product Bible v1.0 and the mandatory feature specification template.
Preview 0.2

Organize and compression results

  • Released private local page previews for Organize PDF.
  • Added reorder, rotate, remove, and extract behavior in one workflow.
  • Added original, result, bytes saved, and percentage saved after compression.
  • Added automatic download and a download-again fallback.
Preview 0.1

Protected processing baseline

  • Released Merge, Split, and Compress PDF.
  • Connected NovaCore tenant, plan, usage, session, and audit contracts.
  • Released the persistent Job API with protected result access and expiry.
  • Added atomic static deployment, health checks, backup, and rollback.