Pruneify Changelog

Full legal document suite (Privacy Policy, Terms, GDPR, Cookie Policy) aligned to actual data flows, plus deep-dive guides on detecting and humanizing AI text.

• feature: Replaced Privacy Policy, Terms of Service, GDPR page, and Cookie Policy with in-depth documents tailored to Pruneify's client-side architecture, Discord contact form, and local browser storage
• improvement: Privacy Explained guide rewritten to accurately describe localStorage/IndexedDB session and history storage, with clear instructions for clearing browser data
• feature: New long-form guides: Humanize AI Text, AI Text Detection Tool, AI Text Humanizer, and How to Make AI Text Undetectable, all optimized for search and linked from the Guide hub and HTML sitemap
• improvement: How to Detect AI-Generated Text and Heuristics Breakdown guides expanded with deeper explanations of detection signals, workflows, and educator/writer use cases
• improvement: Removed outdated JSON Formatting and Export Formats guides and cleaned up internal links to keep the content cluster focused on AI detection and humanization
• improvement: SEO infrastructure tightened: centralized sitemap route config and BASE_URL support in SEO utilities so sitemap, robots.txt, ai.txt, and llms.txt all stay in sync across environments

Per-page Open Graph images with page titles. HTML sitemap at /site-map. Security headers (X-Content-Type-Options, X-Frame-Options, Referrer-Policy) in next.config.

• improvement: Dynamic OG images: root, content routes, and /app each render page-specific titles
• improvement: HTML sitemap at /site-map with Main, Guide, Support, Legal sections
• improvement: Security headers added via next.config.ts headers()

ai.txt expanded with Description, Features, Use Cases, Example Queries for AI crawlers. Placeholder social links removed from Footer.

• improvement: ai.txt: Description, Features, Use Cases, Example Queries sections
• improvement: Footer: removed placeholder GitHub, Twitter, LinkedIn links

Custom branded 404 page with helpful links. Atom feed at /feed.xml for changelog. Footer Compare links to /comparison.

• improvement: Custom 404 page (not-found.tsx) with Nav, Footer, quick links; robots noindex
• improvement: Atom feed at /feed.xml generated from changelog; alternates link in layout
• improvement: Footer Compare link fixed: /app#compare to /comparison (orphan fix)

Organization, WebSite, WebApplication, FAQPage, BreadcrumbList, HowTo, WebPage, Article schemas. llms.txt v1.1.1 Markdown format. Sitemap images, robots rules, favicon, manifest icons.

• improvement: Organization and WebSite JSON-LD in root layout
• improvement: WebApplication schema on /app; BreadcrumbList, WebPage schemas on content pages
• improvement: HowTo schema on /guide/how-to-detect-ai; Article schema on all guide pages
• improvement: llms.txt rewritten to spec v1.1.1: H1, blockquote, H2 sections, annotated links
• improvement: Sitemap images for / and /app; robots.txt rules for ClaudeBot, PerplexityBot, Bingbot, etc.
• improvement: Favicon (icon, apple-icon, 192, 512); manifest icons array for PWA

Canonical URLs for all pages to reduce duplicate content risk. theme-color meta for mobile browser chrome and PWA.

• improvement: alternates.canonical per page; absolute URLs via metadataBase
• improvement: theme-color meta (#00D4FF) for mobile and PWA appearance

Rich previews on Twitter/X with large image cards. twitter:card, twitter:title, twitter:description, twitter:image, twitter:creator.

• improvement: Twitter Card tags: summary_large_image, title, description, image
• improvement: twitter:creator @pruneify for attribution

Rich previews on Facebook, LinkedIn, Slack, Discord. metadataBase, openGraph, dynamic OG image.

• improvement: Open Graph tags: og:title, og:description, og:image, og:type, og:site_name
• improvement: metadataBase for absolute URLs on social platforms
• improvement: Dynamic OG image (1200x630) via opengraph-image.tsx

All page meta descriptions lengthened to 110-160 chars for better SERP visibility. /app layout expanded from 47 to 158 chars. CTAs added.

• improvement: Meta descriptions lengthened to 150-160 chars on key product pages
• improvement: CTAs added: 'Try free', 'Try free today', 'Try free — no signup' where appropriate
• improvement: /app layout description expanded from 47 to 158 chars (was critically short)

All page titles lengthened to 50-60 chars for better SERP visibility. Added title.template for consistent branding across the site.

• improvement: Root layout: title.template ("%s | Pruneify") and default title 50 chars
• improvement: All page titles lengthened to 50-63 chars for optimal search visibility
• improvement: Short titles updated: /app, /privacy, /terms, /resources, /cookie-policy, /gdpr, guides

Comprehensive LLM text cleanup as the first step in every processing flow. Removes invisible chars, normalizes typography, strips AI metadata, and optionally humanizes common AI phrases. Web Worker offload for large texts.

• feature: Ultimate LLM Normalization Engine runs first in pipeline (before detection/formatting)
• feature: Invisible and zero-width char removal (ZWSP, BOM, control chars)
• feature: Typography normalization: em/en dash, curly quotes, ellipsis, NBSP
• feature: AI metadata stripping: oai_citation, grok_card, utm_source, contentReference, etc.
• feature: Optional phrase humanization (delve into → explore, utilize → use, 30+ mappings)
• feature: LLM Cleanup section in Controls: master toggle, Aggressive Dashes, Humanize Phrases, Remove Metadata
• feature: Preset buttons: ChatGPT/Claude/Gemini, Code-Ready, Prose-Friendly
• feature: Cleanup Summary card with green badges and expandable details
• feature: View normalization diff: before/after diff for normalization step
• feature: Web Worker offload for texts >50k chars to avoid blocking main thread
• feature: window.llmNormalizer exposed for console debug and testing
• feature: Replace visible elements: per-toggle for em dash —, en dash –, curly quotes, ellipsis …, Markdown (# * `), Unicode bullets • (choose what to replace)
• improvement: New settings persisted: aggressiveLLMCleanup, aggressiveDashes, preserveProseDashes, humanizePhrases, removeMetadata, replaceEmDash, replaceEnDash, replaceCurlyQuotes, replaceEllipsis, replaceBullets

History stored in IndexedDB (last 50), clickable to load, select-based delete. Compare moved into /app. Format preserves structure. Scroll sync.

• feature: History stored in IndexedDB (last 50 entries persisted in browser)
• feature: History items clickable: loads input and output into formatter
• feature: History: Select all, checkboxes, Delete selected (replaced Delete all)
• feature: History list scrollbar (max-height 220px) to prevent UI overflow
• feature: Format preserves newlines, tabs, and indentation
• feature: Input and output scroll synchronized
• feature: Input and output same fixed height (320px)
• feature: Compare tool moved into /app below Controls; /compare redirects to /app
• feature: Before/After panels scroll synchronized in side-by-side view
• feature: Format after load-from-history updates existing entry (no duplicate)
• improvement: Compare: Removed Two texts mode; centered heading and view buttons
• improvement: Removed Changes button from Output block
• improvement: Removed Signal breakdown from Detection panel
• improvement: Nav and Footer logos use next/image

Major redesign focusing on removing LLM footprints (invisible chars, curly quotes, em-dashes) as the primary value. Formatter and Compare are now separate tools. Full nav and footer added to all tool pages.

• feature: Dedicated Compare tool at /compare with Before/After and Two-text diff modes and LCS-based line diff
• feature: Footprint Log panel showing changes by category (invisible, curly-quotes, dashes, tabs, ellipsis) with expandable details
• feature: One-click Copy cleaned text button in footer with success feedback
• feature: Normalization enhancements: tab normalization, ellipsis (U+2026 to ...), stripEmoji exposed in controls
• feature: Nav and Footer added to /app and /compare so users can access all site links from tool pages
• improvement: Formatter refocus: hero and metadata emphasize footprint removal; Output and Footprint Log are primary
• improvement: Detection de-emphasis: AI detection moved to collapsible Advanced section with optional Show AI detection toggle (default off)
• improvement: Footer spec alignment: Product column (Formatter, Compare), restructured trust strip with heading, Documentation link
• improvement: Input panel simplified: removed Response tabs and Add button; only Paste and Upload remain
• improvement: Nav link hover underline animation and 150ms transitions

First public release of Pruneify, a client-side LLM response detector and formatter with dark UI and scroll-animated landing page.

• feature: Launch of Pruneify landing page and /app analysis workspace

Introduced a JSON-based changelog file and SEO-optimized /changelog page to transparently communicate Pruneify releases and improvements.

• feature: Added structured public/changelog.json file for version tracking
• feature: Created /changelog page rendering entries in newest-first order
• improvement: Added SEO-focused titles, descriptions, and schema markup for release notes