| 开发者 | sumitsingh |
|---|---|
| 更新时间 | 2026年6月20日 23:36 |
| PHP版本: | 7.4 及以上 |
| WordPress版本: | 7.0 |
| 版权: | GPLv2 or later |
| 版权网址: | 版权信息 |
rankbix-seo folder to /wp-content/plugins/.Yes. Rank Bix SEO supports WooCommerce products in the post editor meta box, bulk SEO editor, sitemap, and schema features.
Yes. Rank Bix SEO supports public custom post types with UI. They can be used in the SEO meta box, sitemap, and bulk editor.
Yes. Go to Rank Bix SEO → Content Manager → Import SEO tab and choose your source plugin. Rank Bix SEO can import SEO titles, meta descriptions, focus keywords, canonical URLs, and noindex settings.
No. Rank Bix SEO is designed for performance. The XML sitemap is cached, AI model lists are loaded only when needed, and redirects use a dedicated database table with a fast lookup process.
Yes. You need at least one API key from a supported AI provider. AI features do nothing if no key is configured.
No. Data is only sent when you configure an AI provider API key and manually click an AI generation button in the post editor.
When you click an AI generation button, Rank Bix SEO sends the post title and optionally a plain-text excerpt of the post content to the selected provider. You can disable sending post content in Settings → Advanced → AI Keys → Allow sending post content. When disabled, only the post title is sent.
Yes. Go to Rank Bix SEO → Settings → Sitemap and uncheck Auto-pings Google & Bing on publish.
Add this function to your theme template:
<?php rankbix_seo_breadcrumbs(); ?>
Alternatively, use the shortcode:
[rankbix_seo_breadcrumbs]
Use this shortcode on any page:
[rankbix_seo_sitemap]
Optional shortcode attributes include show_tags="0", columns="3", and orderby="date".
Visit any redirected URL and append ?rankbix_seo_debug_redirect=1. Logged-in administrators can see redirect match details instead of being redirected.
Yes. You can use only the features you need, such as meta titles, descriptions, Open Graph, sitemap, schema, redirects, or breadcrumbs.
wpml_object_id filter for per-language URL resolution; custom field translation for WPML can be configured via WPML → Settings → Custom Fields Translation.<url> entry in the XML sitemap now includes <xhtml:link rel="alternate" hreflang="..."> tags for every available translation, with an additional x-default entry pointing to the default-language version. The xmlns:xhtml namespace declaration is added to <urlset> only when hreflang links are present, keeping single-language sitemaps unchanged. Translated post URLs are injected as their own <url> entries with their own full hreflang set — ensuring Google sees reciprocal hreflang on every language version. Sitemap caches are automatically invalidated when a translation is saved (pll_save_post / wpml_translation_job_saved), so the sitemap always reflects the latest translation state.sitemap_index.xml was incorrectly
outputting a flat <urlset> containing every URL on the site in a single file. It now
outputs a <sitemapindex> listing individual sub-sitemaps per content type
(post-sitemap.xml, page-sitemap.xml, category-sitemap.xml,
post_tag-sitemap.xml), matching the structure search engines expect. The
"← Back to Sitemap Index" link in the browser view now correctly navigates back to the
index instead of reloading the same page. /sitemap.xml now 301-redirects to
/sitemap_index.xml. Trailing-slash variants of sitemap URLs are also redirected to
the canonical form to prevent duplicate indexing.first_seen datetime column to the log table so each URL's discovery date is tracked, enabling the per-day chart.<?xml-stylesheet?> processing instruction pointing to assets/sitemap.xsl. Browsers automatically download and apply the stylesheet, rendering a fully styled HTML page (blue gradient header, branded table layout) — identical in appearance to Rank Math's sitemap view. Search engine crawlers ignore the PI and receive clean XML as before./sitemap_index.xml is now the primary sitemap URL, matching the Pro version. /sitemap.xml continues to work for backwards compatibility. robots.txt and search engine auto-pings now advertise /sitemap_index.xml.[rankbix_seo_toc] shortcode ignoring the custom TOC title saved in the post meta box. The shortcode now uses the _rankbix_seo_toc_title meta value as the default title before falling back to "Table of Contents".#general, #sitemap, #tools) to correctly open the target settings tab.wp_options) and includes a tip for hardening via wp-config.php.Settings → Reading), preventing your site from being silently invisible to Google.?p=123), which produces non-descriptive URLs that hurt crawlability and rankings. Links directly to Settings → Permalinks.%post_title% %sep% %site_name%) going blank after saving. Template options for all post types and taxonomy archives were registered outside of admin_init, so WordPress's Settings API did not recognise them and silently discarded the values on every save. Props @lozmatic for the report.sessionStorage and restored after the options.php redirect, so you land on the tab you were working on.admin.php?page=rankbix-seo-settings#advanced opens the Advanced tab directly. Hash takes priority over the saved tab; useful for linking from documentation or support replies.load_plugin_textdomain() so translation files in the /languages folder are correctly loaded for all string translations.pagespeed.web.dev/?url= to pagespeed.web.dev/analysis?url=.rankbix_seo_onboarding_dismissed and rankbix_seo_version options added in v1.1.1. Fixed physical llms.txt file at the WordPress root not being deleted on plugin uninstall.responseSchema now enforces both title and description keys at the API level, and the prompt explicitly states both fields are required.models/, which doubled the path in the generate URL (.../models/models/gemini-2.0-flash:generateContent). Prefix is now stripped correctly.gemini-1.5-flash to gemini-2.0-flash. Updated meta box quick-select options to gemini-2.0-flash and gemini-2.5-flash.responseMimeType: application/json and responseSchema to Gemini requests — forces clean JSON output with no markdown code fences or explanation text, eliminating the most common parse failure.claude-haiku-4-5-20251001 (fully versioned ID for production stability).llama3-8b-8192 to llama-3.1-8b-instant. Added Groq to the live model list fetch so the retry fallback has real model options./v1/chat API (using message + preamble fields) to the current v2 /v2/chat API (using messages array). Updated response extraction to match the v2 response structure.json_decode fails due to unescaped characters, and shows the full raw response in the error message for easier debugging.maxOutputTokens from 400 to 800 across all providers to prevent response truncation mid-JSON.the_content filter which fires after wp_head(), so the CSS never loaded. Moved enqueueing to the wp action hook so styles are always included in the page <head>.<p class="wp-block-paragraph">). The previous check for a bare <p> missed these, causing the TOC to be prepended outside of the content flow. Now uses offset-based substr insertion that matches any opening paragraph tag.--wp-rankbix-seo-color) with self-contained CSS variables defined inside .rankbix-seo-toc, added card shadow, title separator, display: block layout with max-width: 480px, lower-alpha sub-list style, link hover transition, smooth scroll via html:has(.rankbix-seo-toc), and full-width mobile handling.?url= instead of /analysis?url=), which opened the PageSpeed homepage rather than running an analysis on the current page.rmsOnProviderChange() to update button state. Fixed by calling rmsOnProviderChange() immediately after the dropdown is populated.?key=). Key is now sent via the X-Goog-Api-Key header, keeping it out of server logs and browser history.fill_missing_alt() and fix_img_tag() that would silently overwrite intentionally set alt text on images.rmsImportStatus() using innerHTML with partially controlled content. Replaced with textContent and createElement to prevent potential XSS.s.url into href and onclick attributes. All URL values are now handled via createElement and addEventListener.sanitize_text_field() being applied to $_SERVER['REQUEST_URI'] before URL parsing, which stripped valid percent-encoded characters and caused redirect matching to fail for encoded URLs.current_user_can('manage_options') capability check to render_duplicate_page().suppress_all_notices() calling remove_all_actions('admin_notices'), which silently removed security alerts from other plugins (Wordfence, Jetpack, etc.) on RankBix SEO admin pages.rankbix_seo_duplicate_prefix but the class read rankbix_seo_duplicate_title_prefix.get_year_link(), get_month_link(), and get_day_link().the_content filter replacing all post content on the virtual sitemap page instead of only the sitemap post.RankBixSEO_Robots and RankBixSEO_Sitemap independently appending the sitemap URL.save_post and delete_post.// phpcs:ignoreFile blanket suppressions from class-rankbix-seo-admin.php, uninstall.php, and class-rankbix-seo-importer.php. Replaced with targeted inline suppression comments where genuinely needed.RankBixSEO_Admin::header() and footer() made public static. RankBixSEO_BulkEditor now uses the shared header and footer, eliminating a duplicate nav implementation that was missing the 404 Monitor tab.<form> tag in the Bulk SEO Editor.$raw_from assignment and redundant triple esc_sql() calls in the redirections handler.