Rank Bix SEO – Lightweight SEO Toolkit with Multi-AI

Өзгөртүүлөр

1.2.4

New Features

• New module: AI Content – Write complete, SEO-ready draft posts with your own AI key, one at a time or on a schedule. Find it under Rank Bix > AI Content after turning it on in Features (it is off by default). It has four tabs: Generate, History & logs, Content defaults and Connection test. It uses the API keys you already set in Settings > Tools, so nothing is entered twice and existing keys keep working. The Connection test tab lists each provider with a key, the model it will use and a Test connection button. If a provider has retired the default model, Rank Bix switches to a working model from your account and tells you which one. The Content defaults tab sets which AI writes, which creates images and the image model, your brand (name, what you do, audience, tone, language, words to avoid, pre-filled from your site and Local Business details), article length, number of H2 sections, FAQ count (saved as FAQ schema), default category, and featured and inline images with an image style. A Preview full prompt button shows exactly what will be sent to the AI.
• AI post generation – AI Content now writes complete draft posts. Enter a focus keyword (and optionally a title, or click Suggest titles) and Rank Bix plans the outline, SEO title, meta description, slug, tags, category and FAQs, writes each section, creates the featured and inline images, then saves a draft with native blocks, the Rank Bix SEO fields, FAQ schema and an SEO score. Each part runs as its own small request, so long posts never hit server timeouts, and live progress shows every step.
• AI content queue – Queue up to 50 keywords at a time and Rank Bix creates one draft per interval (every 15 minutes up to once a week) in the background with WordPress cron. Posts are always saved as drafts for review.
• History & logs – Every generated post has a step-by-step log with timings and the model used. Failed steps retry automatically, failed images are skipped without losing the post, and stopped jobs can be retried from where they stopped.
• AI Content safeguards – A new “Maximum AI posts per day” setting (default 20) protects your API bill; queued posts over the limit move to the next day. The queue shows how many drafts and paid image requests it will create before you add it. If WordPress cron is not running, History & logs warns about overdue posts and offers a Run the queue now button. Images from stopped jobs are cleaned up after 7 days, and finished job logs after 90 days.
• AI provider fallback – When the writing or image provider is rate limited, out of quota or overloaded, AI Content switches to another connected provider for that step and logs the switch. Rate limits and overloads now wait as long as the provider asks before retrying.
• Getting Started checklist – A new dashboard card for fresh installs. It detects existing Yoast, Rank Math, AIOSEO or SEOPress data and prompts an import, then checks the homepage SEO title, default social image, XML sitemap and Search Console submission. Steps tick off automatically as they are completed, with a progress bar and a dismiss option.
• “Show me” setup guides – Each Getting Started step has a “Show me” button that opens the right settings page, switches to the right tab and highlights the exact field to fill in, with a short tip on what to enter and where to save.
• Quick tour – A four-stop welcome tour opens once for each admin on their first dashboard visit, pointing out the checklist, Features and Settings. It can be replayed any time from the “Quick tour” button on the dashboard.

Improvements

• Automatic SEO titles and descriptions per post type – Each post type now has a meta description template next to its title template (Settings > General > Title & Description Templates). Posts with an empty SEO Title or Meta Description use them automatically, and like Yoast and Rank Math nothing is saved, so renaming a post or changing a template updates every post that has no custom value. The default description is now a clean summary of up to 155 characters that ends on a full sentence or word (it used to be 30 words, often over Google’s limit). The post editor shows the automatic title and description in grey in the fields and in the search and social previews, updating live as you edit the post title, and the SEO score now checks them too.
• Near-zero queries on visitor pages – Rank Bix now remembers which of its settings a site has never saved and tells WordPress up front, so reading them costs no database query (a setting drops off that list the moment it is saved). Normal redirect lookups, the default social image and the location schema are also cached. On our test sites a Rank Bix homepage now runs 1 to 4 database queries, search pages none, and a 404 page only the 404 log entry.
• Lighter 404 logging – Logging a missing URL no longer rewrites cache entries on every hit, which mattered most when bots request many missing URLs.
• No wasted cache rebuilds – Opening “Add New”, saving drafts and autosaves no longer clear the sitemap, SEO score and Bulk Editor caches. Only changes to published content clear the sitemap.
• Faster front end – Rank Bix now runs far fewer database queries on every page view (on our test site the homepage went from 51 to 9 Rank Bix queries, single posts from 17 to 6, sitemaps from 4 to 1). Image SEO no longer searches the media library for images that already have alt and title text, sites with no redirects run no redirect query, the redirect lookup and its trailing-slash fallback are one query, and settings read on every page are now loaded with WordPress’s other settings instead of one query each. Page output is unchanged.
• Lighter admin – The redirects table check no longer runs on every admin page, and the slug-change notice only looks for its data on post screens.
• Sitemap self-repair – If another plugin or a settings change clears WordPress’s rewrite rules while Rank Bix is not loaded, sitemap_index.xml stopped working until the next update. Rank Bix now notices this on the next admin visit and restores the rules.
• OpenAI-compatible AI provider – A new option in Settings > Tools > AI Meta Generator – API Keys connects any service with an OpenAI-compatible API, including OpenRouter, DeepSeek, Mistral, Together, Fireworks, xAI, Perplexity and local Ollama. Enter a base URL, an optional key and a model name.
• API keys in wp-config.php – AI keys can now be defined as constants (for example RANKBIX_SEO_OPENAI_KEY), which take priority over keys saved in the database.
• Faster, more reliable AI calls – All AI features now share one connection layer. A wrong API key fails straight away instead of retrying several models, model lists are cached for six hours instead of fetched on every editor visit, and retired default Groq models were replaced.
• AI posts built to score 80+ – The SEO title is fitted to the right length with the focus keyword near the start, the URL slug is built from the keyword, and a new “Cite one trusted external source” option adds one outbound link that is checked to exist before it is added.
• Modern AI Content screens – Generate now has one composer with option chips and a live “This post will” summary. The progress timeline shows how long each step took. The queue card previews the publish dates and the number of AI requests before you add it. Content defaults has a side menu and an unsaved-changes bar.
• History & logs filters and pages – Filter jobs by All, Queued, In progress, Done, Stopped or Cancelled, each with a count, 20 per page. Row buttons (Edit draft or Edit post, Log, Retry, Delete) share one design, and an open log is highlighted together with its row.
• Separate text and image limits – When an image provider runs out of quota (for example a free Gemini key with no image quota), only images pause and the text keeps being written. The log says exactly what to fix.

Bug Fixes

• Fixed SEO scores, keyword density, content length and reading time being wrong for non-English content (for example Hindi, Arabic or Russian). Words are now counted correctly in every script.
• Fixed the “Focus keyword in URL slug” check failing for keywords and slugs written in non-Latin scripts.
• Fixed the AI meta generator sending post content even when “Allow sending post content to AI providers” was turned off.
• Fixed the “SEO: Sitemap Off” toolbar alert and the dashboard widget alert linking to the Sitemap settings tab, which cannot turn the sitemap back on. They now open Features with the XML Sitemap toggle highlighted.
• Fixed PHP warnings (“Undefined array key 5”) written to the error log on every visit to the Features page.
• Fixed the redirect count on the dashboard not updating after a CSV import or an automatic slug-change redirect.
• Fixed “Delete all data on uninstall” leaving many Rank Bix settings behind (per post type and taxonomy noindex, breadcrumb and TOC styles, licence details and scan results). Every Rank Bix option and cached value is now removed when this option is ticked. Nothing changes when it is not ticked.
• Fixed noindex and password-protected posts appearing in the multilingual (hreflang) sitemap.
• Fixed Product schema with a single star rating being output as an aggregate rating from 1 review, which Google can flag as invalid. It is now a Review with the post author, as Google requires.

1.2.3

Compatibility

• Tested and confirmed compatible with WordPress 7.1.

New Features

• Header & Footer code injector – A new Technical module that lets you inject custom scripts, styles, and tracking codes (Google Analytics, Tag Manager, cookie banners, Meta Pixel, etc.) into the <head>, immediately after <body> open, or before </body> close on every page. Three independent CodeMirror HTML editors with real-time syntax highlighting. The Save button disables automatically when syntax errors are detected. PHP tags are stripped server-side so only HTML runs. The module is zero-impact when toggled off – no frontend output, no admin menu entry.
• Conditional injection – Each of the three Header & Footer code areas (head, body, footer) now has an “Apply to” dropdown so you can target injection to All pages, Homepage only, All singular pages, Blog posts only, or Pages only. Mix and match per area – for example, inject Analytics globally but target a cookie banner to all singular pages only.
• Per-post disable – A new checkbox in the SEO meta box Advanced tab lets you disable all head/footer code injection on individual posts or pages. Useful for landing pages or pages with their own tracking setup.
• Dashboard Module Status – Header & Footer now appears in the Module Status card on the dashboard so you can see its on/off state at a glance alongside all other modules.
• Admin toolbar – Header & Footer is now listed in the Rank Bix SEO admin toolbar dropdown for quick one-click access from any page.
• Hide toolbar score badge – A new option in Settings > General > Admin Toolbar lets you hide the coloured SEO score pill from the toolbar. Useful on PHP-template sites where the score count may not reflect actual frontend output.
• Toolbar score badge – smart threshold – The SEO score badge in the admin toolbar is now hidden automatically when the stored post content has fewer than 50 words after stripping shortcodes and HTML tags. Prevents a misleading score from appearing on PHP-template or page-builder pages where the editor content is minimal but the rendered frontend output is not.
• Keyword in URL slug – skip for homepage and posts page – The “Focus keyword in URL slug” check is now skipped entirely when editing the static front page or the designated blog archive page. The slug for those pages does not represent their public URL (which is always /), so a failing slug check was misleading and is no longer shown.
• SEOPress importer – You can now import titles, descriptions, focus keywords, canonicals, noindex, and nofollow settings from SEOPress. Appears automatically on the Import tab whenever SEOPress data is detected in the database.
• Cookie Bar – A new Technical module for GDPR-compliant cookie consent. Displays a fully customisable banner (position, message, button text, background/text colours, cookie duration). Integrates with Google Consent Mode v2 and Matomo. Automatically blocks scripts injected via the Header & Footer module until the visitor accepts. Toggling the module off disables all frontend output and removes the admin menu entry.

1.2.2

New Features

• Header & Footer code injector – A new Technical module that lets you inject custom scripts, styles, and tracking codes (Google Analytics, Tag Manager, cookie banners, Meta Pixel, etc.) into the <head>, immediately after <body> open, or before </body> close on every page. Three independent CodeMirror HTML editors with real-time syntax highlighting. The Save button disables automatically when syntax errors are detected. PHP tags are stripped server-side so only HTML runs. The module is zero-impact when toggled off – no frontend output, no admin menu entry.

1.2.1

New Features

• Redesigned SEO meta box – The SEO tab has been completely rebuilt with a modern, data-rich layout that makes it faster to optimise every post without switching tools. The redesign is non-destructive – all saved SEO data is preserved and works without any migration.
• Live SEO score ring – An animated circular score ring sits at the top of the SEO tab showing a 0-100 score based on your keyword analysis results. The ring fills with a colour-coded stroke (green 70+, amber 40-69, red below 40) and updates live after every analysis run. A “Great / Good / Needs work” label appears beneath the number.
• Keyword density chip – As you type your focus keyword, a real-time KD badge appears showing the keyword density as a percentage of total post content. Green for the ideal 0.5-2.5% range, amber outside it.
• Keyword-in-field indicators – Green “Keyword found” or red “Keyword missing” badges appear next to the SEO Title and Meta Description labels and update live as you type in any of the three fields – no need to run a full analysis to see whether your keyword is present.
• Visual length bars – SEO Title and Meta Description each show a colour-coded progress bar beneath the input reflecting the current character count relative to the recommended range (green = good, amber = short, red = over).
• Content stats strip – Word count, estimated reading time (200 wpm), and internal/external link counts are calculated from the post content and displayed in the keyword header on every page load. Word count is a feature many competing plugins lock behind premium plans.
• Search Preview open by default with device toggle – The SERP preview now expands automatically when the meta box loads. A Desktop/Mobile toggle in the header lets you switch between a full desktop result and a mobile card view without leaving the tab.
• Featured image in SERP preview – The desktop and mobile SERP preview cards now show a thumbnail of the post’s featured image (or OG image if set) on the right side of the result snippet, matching how Google displays image-rich results.
• SEO Title column – The posts list now shows the full SEO title (up to 60 characters) giving you a clear at-a-glance view of your optimised title without opening each post.
• SEO Desc column – The posts list now shows the meta description (up to 160 characters), letting you review your full description content at a glance across all posts.

Bug Fixes

• Social preview cards showing “No image selected” despite featured image set – On block editor (Gutenberg) pages, the social preview cards incorrectly showed the placeholder text on page load even when a featured image was set. The JS initialisation ran at 50ms before Gutenberg’s media store cached the attachment, found no image, and overwrote the PHP-rendered card. Fixed by reading the image URL from the PHP-rendered preview element (which already has the correct URL resolved server-side) rather than racing the async media store.
• OG image Remove button not clearing SERP thumbnail – Clicking Remove on the OG image left a stale URL in the preview image element, which caused the old image to reappear when the social cards next updated. The Remove handler now clears the src attribute before calling the image update routine.

Improvements

• SEO Title optimal range updated to 50-60 characters – The recommended title length has been tightened from 40-70 to 50-60 characters, matching Google’s desktop truncation point (~580-600 px). The progress bar, character counter, score check, bulk editor, and all JS now reflect the new range consistently.
• Post and Page are always enabled for SEO Meta Box – Post and Page post types are permanently enabled in Settings and cannot be disabled, removing a common source of misconfiguration. All other post types remain optional.
• Default post types changed to Post and Page – Fresh installs and fallback behaviour now default to Post and Page only, rather than all public post types on the site.
• Focus Keyword column removed from posts list – The Focus Keyword column has been removed from the posts/pages list table to reduce clutter. The keyword is still used internally for all score and analysis calculations.
• TOC theming via CSS custom properties – The Table of Contents block now uses CSS custom properties (--toc-max-width, --toc-title-size, --toc-item-size) for its layout and sizing, making site-level theme overrides simpler without needing to target deeply nested selectors.

1.2.0

New: Complete UI Redesign with Sidebar Navigation

• New sidebar navigation – All plugin screens now share a persistent left-hand sidebar, giving you one-click access to Dashboard, Settings, Image SEO, Redirections, 404 Monitor, Schema Templates, Rank Tracker, and all other modules from any page
• Dashboard SEO Scores widget – A new SEO Scores card on the dashboard shows a Good / OK / Poor breakdown of your published content, with a post-type switcher (Posts, Pages, and any custom post types) and direct View links that open the post list pre-filtered by score band
• Settings tabs reorganised – Image SEO moved from the Advanced tab to General (where always-on core features live); AI Meta Generator API keys and model preferences moved from Advanced to Tools; 404 Monitor removed from Settings entirely (free) – access it directly via the sidebar link
• Image SEO template fix – The %filename% variable now correctly resolves the original upload filename on newly uploaded images (previously fell back to the post title due to a WordPress hook timing issue where _wp_attached_file meta is written after add_attachment fires)

1.1.15

• Added: Sitemap post-type exclusions – choose which post types to include or exclude from your XML sitemap
• Added: Split archive noindex controls – separately control noindex for author archives, date archives, and media attachment pages
• Added: Organization / Person Knowledge Graph schema – output JSON-LD on your homepage to power Google Knowledge Panels
• Added: Global robots snippet controls – set max-snippet, max-image-preview, and max-video-preview directives site-wide
• Added: Facebook App ID and Twitter/X site handle settings for Open Graph and Twitter Cards
• Fixed: Duplicate redirect entries when the same source URL is saved more than once
• Fixed: Auto-created redirects causing double-prefix URLs on subdirectory installs
• Fixed: Bulk delete on Redirections page returning “No IDs provided” error
• Fixed: 404 Monitor prune lock now set before the COUNT query to prevent excessive database queries
• Fixed: Pagination offset computed before page clamp, causing empty results on out-of-range page numbers

1.1.14

New Features

• Breadcrumbs module – complete rewrite – The breadcrumb trail builder has been rebuilt from the ground up as a shared static method (build_trail()) used by both the visual renderer and JSON-LD schema output, ensuring they never drift out of sync. The trail now handles WooCommerce products (shop page + deepest product category ancestors), standard blog posts (blog archive page + category ancestors), hierarchical pages (all page ancestors), custom post types (archive link + first hierarchical taxonomy path), taxonomy archives (term ancestors for nested categories/tags), date archives (year, month, day with linked parent levels), search results, 404 pages, and author archives. All breadcrumb links include aria-current="page" on the final item for accessibility and Google compliance.
• Breadcrumbs – dedicated Settings tab – A new Breadcrumbs tab appears in Settings when the module is active. It is automatically hidden when the module is disabled and the settings page redirects to General. The tab provides Display Settings (separator field with quick-pick buttons for common characters, home label, prefix text, include archive page toggle) and a Usage card with the PHP function and shortcode reference.

Improvements

• Settings – Robots & Indexing tab reordered – The robots.txt editor and LLMs.txt cards now appear at the top of the tab, with the noindex settings card below. Files that are actively served take visual priority over indexing directives.
• Features page – Configure links – All modules on the Features page now show a Configure link pointing directly to their relevant settings section: Breadcrumbs to #breadcrumbs, IndexNow to #tools, LLMs.txt to #robots, Robots to #robots. Modules with no dedicated settings section show an Active badge only.
• Tools tab – IndexNow key file link – The “Key file served at: /hash.txt” label has been replaced with “Key file: View key file” to avoid repeating the key hash that is already shown above.
• Bulk Editor – private posts excluded – Private posts and non-navigable post types (Elementor templates, JetEngine templates, etc.) are now excluded from all Bulk Editor queries, the post count, the status dropdown, and the Average SEO Score calculation.
• Breadcrumb CSS – CSS custom properties – Frontend styles now use CSS custom properties (--rbs-bc-link, --rbs-bc-current, --rbs-bc-sep-color, --rbs-bc-size) with defaults, making Pro style overrides possible without modifying the stylesheet. Font size unified to 16px (14px on mobile). Elementor kit override compatibility added via !important on color and font-size declarations.
• Meta title – archive pages – Post type archives, date archives (year/month/day), and author archives now generate proper titles. Previously post type archive pages fell through to the else branch returning only the site name.
• Meta title – template variable resolution – %post_title% in a taxonomy title template now resolves to the term title instead of showing literally when called without a post object. All remaining unresolved %variable% placeholders are automatically stripped after replacement so they never appear in the browser tab.
• Meta description – archive and search pages – Meta descriptions are now generated for post type archives (using the post type description), author archives (using the author bio), and search result pages (a “Search results for: query” string). Previously these pages output no description at all.

Bug Fixes

• Error with Yoast SEO – RankBixSEO_TOC::add_heading_ids() and maybe_auto_insert() declared a strict string type on the $content parameter. Yoast SEO calls get_the_excerpt() internally which triggers the the_content filter with null, causing a fatal TypeError. Both methods now accept ?string (nullable) and return early safely.
• Title template %post_title% not replaced on archives – When a taxonomy title template was saved containing %post_title%, it was passed through resolve_template() with a null post object, leaving the placeholder literal in the browser title.
• Post type archives showing only site name – get_title() had no is_post_type_archive() branch, so post type archive pages returned only get_bloginfo('name') instead of the archive title.
• Dashboard widget – false homepage title alert on static front page – The “Homepage SEO title is not set” alert in the dashboard widget checked only the global rankbix_seo_home_title option. When WordPress is configured to use a static page as the front page, the SEO title lives in the page’s own post meta (_rankbix_seo_title), not the global option. The widget now checks post meta for static front pages, suppressing the alert when the title is set on the page itself. The alert link also now points directly to the page editor rather than General settings.

1.1.13

New Features

• Elementor SEO panel – A dedicated SEO tab now appears in the Elementor editor sidebar alongside the Elements and Globals tabs. The panel includes a live Google SERP preview card (showing the post’s OG image or featured image thumbnail, URL breadcrumb, title, and description – all updating in real time as you type), an SEO Title field, a Meta Description field, a Focus Keyword field, and an Advanced tab with Canonical URL, Robots directives (noindex, nofollow, noarchive, nosnippet), and a Cornerstone Content flag. All fields save automatically when you click Elementor’s own Update button – no separate save action needed. The panel is hidden for Elementor template post types (elementor_library, landing pages, floating buttons) where per-post SEO data is not applicable. If an AI provider key is configured in Settings, a Generate button appears next to the SEO Title and Meta Description fields to produce AI-written copy using the post title, content, and focus keyword.
• Bulk Editor – AI generation – A “Generate SEO for selected” button appears in the Bulk Editor toolbar when one or more rows are checked. Clicking it processes each selected row in sequence, filling any empty SEO Title, Meta Description, and Focus Keyword fields. If an AI provider key is configured it generates copy using the post title and cleaned page content; otherwise it derives values from the post title and first paragraph. Existing field values are never overwritten. Each processed row pulses blue while in progress, turns green when filled by AI, or yellow when filled from page content, so progress is visible throughout.

Improvements

• Character count colour logic – The character progress bar and counter on SEO Title (optimal 30-60 chars) and Meta Description (optimal 120-160 chars) fields now use three distinct states: yellow when the text is too short to be effective, green when it is within the recommended range, and red when it exceeds the limit. Previously the bar turned green as soon as any character was entered, making it impossible to tell whether the length was actually useful.
• Homepage SEO – static front page notice – When WordPress is configured to use a static page as the front page (Settings > Reading), the Homepage SEO section in the plugin’s General settings now displays a notice explaining that the page’s SEO fields should be edited directly on that page, with a button to open it for editing. Only the Title Separator field remains visible in this state, as it applies site-wide rather than to the front page specifically.
• Bulk Editor – per-page 30-row option – The rows-per-page selector now includes a 30-row option alongside the existing 20, 50, 100, and 200 options. The default is 20 rows per page.
• AI content cleaning – Shortcodes and excess whitespace are stripped from post content before it is sent to an AI provider or used for content-based fallback generation. This prevents raw shortcode tags such as [pfc id="62"] from appearing in generated copy.
• WordPress core sitemap suppressed – When the Rank Bix SEO sitemap module is active, WordPress’s built-in sitemap (/wp-sitemap.xml) is now automatically disabled. Having two sitemaps active at once causes duplicate URLs and confuses search engine crawlers. If the sitemap module is turned off, the WordPress default comes back as normal.

Bug Fixes

• Noindex checkbox out of sync between tabs – The meta box has a “Noindex this page” checkbox on the SEO tab and a separate noindex checkbox in the Robots grid on the Advanced tab. Both write to the same meta key but were independent in the UI – unchecking one while the other (on a hidden tab) remained checked meant the meta was saved as noindex even though the user had explicitly unchecked it. Both checkboxes now sync instantly when either is changed.

1.1.12

Bug Fixes

• Duplicate Page – page builder CSS regeneration – When duplicating a page built with Elementor, Beaver Builder, Bricks, Oxygen, or Divi, the builder’s cached CSS (tied to the original post ID) was copied to the duplicate. The builder then skipped regeneration and served stale or mismatched styles on the new post. The CSS cache meta keys (_elementor_css, _fl_builder_cache_ver, _bricks_page_styles, ct_css_style, _et_pb_css) are now excluded from the duplicate copy, forcing each builder to generate clean, post-ID-correct CSS on first open.

For the full changelog see rankbix.com/changelog/.