Download the PHP package mage2kishan/module-advanced-seo without Composer

Magento 2 Advanced SEO

Panth Advanced SEO (Panth_AdvancedSEO) is the core module of the Panth SEO suite for Magento 2. It handles meta titles and descriptions (templates, rules, per-entity overrides, bulk editing), canonical URLs, robots meta, pagination meta, SEO scoring, duplicate meta detection, a crawl audit, product feeds, Google Analytics 4, Google Search Console verification and the Google Indexing API. It is intended for merchants, SEO staff and developers who manage SEO across products, categories and CMS pages. Storefront output is server-rendered through layout XML, blocks and view models, and the module ships a view/frontend/hyva.xml compatibility marker, so it works on Hyva and Luma themes.

Other SEO features (structured data, social meta, hreflang, filter URLs, sitemaps, redirects, robots.txt, llms.txt and more) are separate packages that this module lists as Composer suggestions; see Included packages.

Product page: Advanced SEO Extension for Magento 2

Features

Meta templates and tokens

• Meta templates per entity type (product, category, CMS) and store (a template is only used for entities that match its conditions), with patterns for meta title, meta description, meta keywords, SEO name, OG title, OG description, OG image, Twitter card type, robots and URL key, plus priority, conditions and an active flag.
• Token syntax {{token}}, with {{attribute:code}} for any attribute and filters |truncate:N, |title, |strip, |default:value, |upper and |lower.
• Registered tokens: name, price, sku, category, store, attribute, page, description, short_description, filter, url_key, seo_name, category_seo_name, brand, search_query, landing, random.
• Three default templates are installed (Default Product Template, Default Category Template, Default CMS Template).
• Admin template grid with preview, apply, CSV export, mass delete and filters (name, entity type, store view, active); the store view column shows store names and "All Store Views". Saving, applying or deleting templates and rules refreshes the full page cache. Apply keeps the indexed meta of entities that have their own meta title or description (unless Force Template Over Existing Meta is Yes). Templates with cron auto-apply enabled are applied by cron and written to the resolved meta table.
• Keyword search on the Templates (name, meta title, entity type, URL key template), Rules (name, entity type), Custom Canonicals (target URL, entity types), Feeds (name, file name, feed type) and Crawl Results (URL, meta title, canonical) grids.
• Title and description length limits (the <meta name="title"> tag always matches the <title>), fallback descriptions built from the short description or category description without inline style, script or Page Builder CSS, optional store name suffix, optional removal of Magento's native title prefix/suffix, and a configurable page-number label on paginated listings.
• CMS pages whose identifier starts with landing-, or whose layout update contains the landing_page handle, are treated as landing pages. Category listings filtered by the brand attribute configured in Panth Structured Data are treated as brand pages.
• Search result pages get their own meta and NOINDEX,FOLLOW (a search template can set other robots), so they also get no canonical tag while Disable Canonical for NOINDEX Pages is Yes; otherwise the canonical keeps the q parameter. The 404 page gets noindex,follow and no canonical (configurable).

SEO rules engine

• Rules per entity type and store with priority, active flag and "stop processing subsequent rules". Rules are evaluated against the entity on every page view, so a new or changed rule applies without a reindex.
• Conditions: nested all/any groups, attribute comparisons (==, !=, >, >=, <, <=, contains, ncontains, in, nin, including category_ids) and stock status.
• Actions: apply a meta template, set a canonical URL, or set noindex.

Per-entity SEO fields in admin

• Product form: Meta Robots, Custom Canonical URL, OG Title, OG Description, OG Image URL, plus an "SEO Canonical" section.
• Category form: Meta Robots, OG Title, OG Description, OG Image URL, canonical settings. The "Exclude from XML Sitemap" field comes from mage2kishan/module-xml-sitemap.
• CMS page form: Meta Robots, Hreflang Identifier, canonical settings.
• SERP preview and SEO score widget on product and category edit forms.
• EAV attributes added: seo_name (product and category), meta_robots (product and category) and layered_navigation_canonical (product attribute option with values Use Global Setting, Base Category URL, Filtered Page URL, Set NOINDEX).

Bulk Meta Editor and Missing Meta Report

• Admin grid with product, category and CMS page tabs and inline editing of meta title and meta description, with a store view scope switcher. Only the edited values are written, in the selected scope, and the storefront page is refreshed.
• Missing Meta Report listing products and categories without meta title or description (the SEO Dashboard cards use the same default-scope count), with the same store view switcher. The grid is paged on the server (50 rows per page by default, 20 to 200 selectable) and Export CSV downloads every matching row.

Canonical URLs

• Emits <link rel="canonical"> and a Link: <url>; rel="canonical" HTTP header on full HTML storefront pages. The header is read from the canonical tag of the rendered page head; JSON, AJAX and other non-HTML responses are left untouched, and an existing Link header (for example preconnect hints) is kept.
• Query string stripping (all, or a list of parameter names), lowercase host, trailing slash handling (separately for the homepage), page-1 canonical for paginated pages, parent canonical for simple products used in configurable, grouped or bundle products, cross-domain canonical store, ignored paths, and no canonical on noindex pages.
• Product canonical type: without category path, shortest category path or longest category path. Optional shortest category URL path.
• Custom Canonicals grid: map a source product, category or CMS page to a target URL or entity per store. The target URL must be a full http:// or https:// URL or a path that starts with /; it may be left empty when a target entity is chosen. A rejected save keeps the entered values, and saving or deleting a custom canonical refreshes the full page cache of the source page.
• Sets Magento's catalog/seo/category_canonical_tag and catalog/seo/product_canonical_tag defaults to 0 and suppresses the native canonical at runtime while the module is enabled, so only one canonical tag is output. With Enable Module set to No (and Emit Canonical Tag Yes) the native product and category canonical tags are output instead, without changing the stored configuration.
• rel="prev" and rel="next" links on paginated category and search result listings (tracking parameters from Strip Query Parameters, or the utm, gclid and fbclid defaults, are left out of these links). The page count comes from the rendered product list (collection size and the current limit), so rel="next" is only printed when a next page exists, and no links are printed for a page number beyond the last page.

URL keys

• Optional URL key generation from a template for new products (and optionally on every save of existing products).
• Regenerates a product's URL rewrites when its visibility changes, and deletes them when it becomes Not Visible Individually.
• Admin action and CLI command to regenerate catalog URL rewrites.

SEO scoring, audit and duplicates

• Per-entity score 0 to 100 with grade A (90+), B (80+), C (70+), D (60+) or F. Weighted checks: length (2.0), duplicate (2.5), readability (1.5), entity (2.0), keyword (2.0) and an "AI overview" content check (1.5). All checks run locally.
• Scores are computed on save (directly or through the message queue) and refreshed by cron.
• Weekly duplicate scan of resolved meta using SHA-256 content hashes.
• SEO Audit page with low-scoring entities, near-duplicate meta and the last crawl summary.
• Crawl audit that fetches storefront pages and records status code, title, description, canonical, robots and issues (missing title, missing description, title too long, description too long, missing canonical, duplicate titles, noindex pages, redirects, redirect chains, fetch errors). Run Crawl in admin queues the crawl and a dedicated cron group runs it, with live progress and a Stop Crawl button. Results appear in the Crawl Results grid.
• Internal link graph with PageRank and related-link suggestions (panth:seo:pagerank).
• Deleting a product, category or CMS page removes its score, embedding, resolved meta and link-graph rows; a daily cron job and panth:seo:prune remove any left behind.

Storefront SEO toolbar

• Optional on-page diagnostics toolbar for allowed IPs (or developer mode only when no IPs are set), with copy canonical/title, view source and links to Google Rich Results Test, opengraph.xyz and the Twitter card validator. Pages that show the toolbar are never stored in the full page cache (the layout is marked not cacheable and the cache key varies), so pages cached for other visitors stay without it; an allowed visitor whose first request is served from the cache sees the toolbar from the next page on.

Product feeds

• Feed profiles (one store view each; a rejected save keeps the entered values) (Google Shopping, Facebook Product Catalog, Custom XML, Custom CSV) with XML (RSS 2.0) or CSV output, optional ZIP or GZIP compression, field mapping (product attribute, static value, template token or parent attribute), category and attribute set filters, stock/status/visibility filters, currency, UTM parameters and optional FTP or SFTP delivery with a connection test. Feed file names may only contain letters, digits, dots, dashes and underscores and must end in .xml, .csv, .txt or .tsv. A default "Google Shopping Feed" profile is installed for store ID 1.
• Legacy Google Merchant feed per store built from the Google Merchant Feed settings.
• Feeds are served at /panth_seo/feed/google (legacy feed) and /panth_seo/feed/google?id=<feed id> (profile feed) when the Google Merchant feed is enabled.

Google Analytics 4 and Search Console

• GA4 gtag.js snippet with view_item and view_item_list events, plus a global helper for add_to_cart.
• google-site-verification meta tag.
• Google Indexing API submission of saved product, category and CMS page URLs.

Performance

• panth_seo_resolved_meta indexer (with mview subscriptions on products, categories, stock items and CMS pages) that precomputes resolved meta. It only rewrites its own columns of panth_seo_resolved, so hreflang, Open Graph and JSON-LD data stored there by other Panth SEO modules is kept.
• Optional asynchronous scoring through the panth_seo.score_entity queue.
• Optional Last-Modified and ETag headers on product and category pages. The ETag is computed from the rendered page, so a 304 Not Modified is only sent for If-None-Match with the ETag of the same page content (another store view, page number, filter or a changed price gets a full response); If-Modified-Since alone always gets the full page.
• Optional Speculation Rules JSON: prerender for .html links and prefetch for other links, both with moderate eagerness, excluding checkout, customer, wishlist and cart paths. Skipped on Hyva store views, whose theme outputs its own rules.

Site Audit (Semrush-style checks)

Every crawl (Run Crawl, the scheduled audit, panth:seo:crawl and panth:seo:audit --format=...) also runs the Site Audit checks on the crawled pages. The checks use the store's own crawl and need no external API. Each issue has a code, a severity (error, warning or notice), a category (crawlability, content, meta, structured_data, links, performance, international, ai_search), the affected URL, the offending element and a "How to fix" text.

How it works:

• Each HTML page is parsed once with DOMDocument; every check receives the parsed page (title, meta, H1s, canonical, robots, hreflang, links, images, JSON-LD with @graph flattened, scripts, styles, text size and landmarks).
• Internal images, CSS, JS and linked files are HEAD-checked once per URL. External links are HEAD-checked with GET as fallback, 8 requests in parallel, 5 s timeout and one retry, capped by Max External Checks per Crawl. 403, 405, 429 and 999 answers from linkedin.com, x.com, twitter.com, facebook.com, instagram.com, medium.com, wa.me and whatsapp.com are unverifiable, not broken.
• robots.txt, llms.txt, the XML sitemap (the Sitemap: lines of robots.txt; without them the file name and path of the active XML sitemap profiles of the store view, and /sitemap.xml when there is no profile; at most Sitemap Sample URLs per sub-sitemap, reported as "sampled"), HSTS headers of the store host and linked subdomains, and a cache-busted fetch of the first pages for load time are read once per crawl.
• The crawler sends User-Agent: PanthSeoAudit/<version> (+<store url>), waits at least 100 ms between requests, stops after 3 consecutive 5xx answers, never sends a session cookie and never crawls /customer, /checkout, /sales, /wishlist, /newsletter, /admin* or URLs with ___SID.
• Checks are classes implementing Panth\AdvancedSEO\Model\Audit\Check\PageCheckInterface (per page) or SiteCheckInterface (whole crawl), registered in the checks argument of Panth\AdvancedSEO\Model\Audit\SiteAuditor in etc/di.xml. Another module can add a check with a di.xml item and describe its code through the definitions argument of Panth\AdvancedSEO\Model\Audit\IssueCatalog.
• Issues are saved per run in panth_seo_audit_issue; the last runs (Issue Runs to Keep) stay for the comparison. The crawl summary also stores the counts by severity.

The legacy crawl checks (missing title and description, length limits, missing canonical, 404, 5xx, fetch errors, redirects and redirect chains) are reported with the same structure; their summary keys are unchanged.

Admin: the SEO Audit page shows the issues of the last run in tabs for Errors, Warnings and Notices with category filters. Each issue type shows "N pages" and expands to the URL, element and detail of every row, has a "How to fix" tooltip, and a "Compare with previous run" column (new, fixed, unchanged). Export CSV downloads the whole run.

CLI: bin/magento panth:seo:audit --store=1 --max-pages=500 --format=table|json|csv --severity=error crawls the store, runs the checks and exits with code 1 when any error-severity issue exists (2 when the audit could not run), so it can gate CI. --severity limits the listing to that severity or worse; the exit code always considers every issue. --no-save skips saving. Without --format, --max-pages or --severity the command prints the resolved-meta JSON summary as before. Inside a container, set PANTH_SEO_CRAWL_INTERNAL_HOST (for example http://nginx or https://host.docker.internal) when the store URL is not reachable from PHP; the original Host header is sent, plus X-Forwarded-Proto: https when an https store is reached over http. With an https override on a local certificate, set Verify TLS Certificate While Crawling to No.

Opt-in fixes (Stores > Configuration > Panth Advanced SEO > Site Audit Fixes, all off by default):

• Strip Invalid Schema Properties (panth_seo/audit_fixes/strip_invalid_schema_props): removes properties not allowed for the node's @type from the Panth_StructuredData provider output.
• Rewrite Relative README Links (panth_seo/audit_fixes/rewrite_relative_readme_links): rewrites href="docs/..." and href="./docs/..." in product and CMS content, at render time, to the raw.githubusercontent.com base of the image inside the link.
• Hreflang Group Guard (panth_seo/audit_fixes/hreflang_group_guard): refuses to save a Panth_Hreflang group with two members for the same locale.
• SEO Title Wins Over H1 (panth_seo/audit_fixes/seo_title_wins_over_h1): re-applies the resolved SEO title after the layout is built so a later title change to the H1 heading does not replace it; the H1 keeps the heading.

After fixing issues on a site behind a CDN, purge the CDN cache before the next crawl, or the audit still sees the old edge copies.

Included packages

composer.json requires one other mage2kishan package:

The following packages are listed under suggest in composer.json. They are not installed automatically; install the ones you need. Descriptions are taken from composer.json.

hyva-themes/magento2-default-theme is also suggested for Hyva frontend use.

Panth Advanced SEO itself does not output JSON-LD, OpenGraph/Twitter tags, hreflang links, filter URLs, sitemaps, redirects, robots.txt or llms.txt; those come from the packages above. It reads the brand attribute and default brand from the Panth Structured Data configuration (panth_structured_data/structured_data/*, with a fallback to panth_seo/structured_data/*), and its SEO Dashboard shows counts for filter rewrites, hreflang groups and sitemap profiles when those packages are installed.

Compatibility

Magento constraints in composer.json: magento/framework ^103.0, magento/module-store ^101.1, magento/module-catalog ^104.0, magento/module-cms ^104.0, magento/module-url-rewrite ^102.0, magento/module-catalog-url-rewrite ^100.4, magento/module-sitemap ^100.4, magento/module-backend ^102.0, magento/module-ui ^101.2, magento/module-config ^101.2, magento/module-eav ^102.1, magento/module-configurable-product ^100.4, magento/module-grouped-product ^100.4, magento/module-bundle ^101.0, magento/module-catalog-inventory ^100.4, magento/module-directory ^100.4.

Requirements

• Magento Open Source or Adobe Commerce 2.4.4 to 2.4.8
• PHP 8.1 to 8.4
• mage2kishan/module-core (installed by Composer)
• Magento cron running (scoring, feeds, duplicate scan, cleanup and the crawl audit run from cron)
• For asynchronous scoring: the panth_seo.score_entity.consumer queue consumer (database connection)
• For the Google Indexing API: a Google Cloud service account with Indexing API access

Installation

setup:di:compile is only needed in production mode. setup:static-content:deploy -f deploys the admin CSS, JavaScript and templates under view/adminhtml/web.

Check the module status:

If you install any of the suggested packages, enable their modules as well.

Configuration

Admin path: Stores > Configuration > Panth Extensions > Advanced SEO (section panth_seo). The "Panth Extensions" tab is defined by Panth_Core. Settings can be set at default, website and store view scope unless noted.

General

Meta Tags

Canonical URLs

URL Key Automation

Reports & Diagnostics

Site Audit Checks

Site Audit Fixes (opt-in)

Google Merchant Feed

All fields except the first are shown only when the feed is enabled.

Google Analytics 4

Google Search Console

Advanced

Usage

Admin pages

Under Panth Infotech > Advanced SEO in the admin menu: SEO Dashboard, Meta Templates, SEO Rules, Bulk Meta Editor, Custom Canonicals, Product Feeds, Missing Meta Report, SEO Audit, Crawl Results and Configuration.

• Meta Templates: create a template, pick entity type and store, enter patterns, optionally enable cron auto-apply, check the saved patterns in the Preview section (up to 10 matching products or categories, nothing is written) and use Apply to write it now.
• SEO Rules: build conditions and choose a template, canonical or noindex action.
• Bulk Meta Editor: switch entity type and store view, edit titles and descriptions inline, then save.
• Custom Canonicals: map a source entity to a target URL or entity.
• Product Feeds: manage profiles and field mappings, add the Google default fields, generate a feed and test FTP/SFTP delivery.
• SEO Audit: review low scores and duplicates, start or stop a crawl, and review the Site Audit issues of the last run (tabs per severity, category filters, "N pages" expansion, How to fix, comparison with the previous run, Export CSV). Crawl Results lists each crawled URL.

Actions that change data (delete a template, rule, custom canonical, feed profile or feed field, Apply a template, Generate a feed, Add Google Defaults, Test FTP Connection) are sent as POST requests with the admin form key. Opening their URLs with GET returns 404.

Storefront output

With the module enabled, pages get the resolved meta title, description and robots, the canonical tag and Link header, rel prev/next on paginated listings and, when enabled, GA4, the verification tag, Speculation Rules, the Last-Modified/ETag headers and the SEO toolbar.

Cron jobs

The panth_advancedseo cron group runs in a separate process.

Console commands

Feed output

Profile feeds are written to pub/media/panth_seo/feeds/. The legacy feed is written to pub/media/panth_seo/google_feed_<store_code>.xml.

External services

• Google Indexing API (only when Enable Indexing API is Yes and a service account key is set): after a product, category or CMS page is saved, at the end of that request the module requests an OAuth token from oauth2.googleapis.com with a JWT signed by the service account key, then sends the entity URL with type URL_UPDATED to indexing.googleapis.com. See the Indexing API quickstart.
• External link checks (only during a crawl, when Check External Links is Yes): the server sends HEAD requests (GET as fallback) with the PanthSeoAudit user agent to the external URLs linked from the crawled pages, up to Max External Checks per Crawl. No store data is sent. Each external host is resolved once; hosts that do not resolve or that resolve to a private or reserved address are reported as failed and not requested, and the request is pinned to the checked IP address so a second DNS answer cannot redirect it to an internal address. Redirects to private or reserved addresses are aborted.
• Google Analytics 4 (only when Enable GA4 is Yes): the visitor's browser loads gtag.js from Google and sends page views and ecommerce events. See the GA4 documentation.
• FTP/SFTP (only for feed profiles with delivery enabled): the generated feed file is uploaded to the configured host. Hosts that do not resolve to an IPv4 address, or that resolve to private or reserved addresses, are rejected, and the connection is made to the checked IP address. Test FTP Connection uses the values in the form and the saved password when the password field is empty. Delivery passwords are stored encrypted, are not sent back to the edit form, and are kept when the password field is left empty on save.
• Crawl audit: requests go to the store's own base URLs. The PANTH_SEO_CRAWL_INTERNAL_HOST environment variable can point the crawler to an internal host (for example inside containers).

No external AI service is called by this module; the scoring checks run locally.

Developer Notes

• Module: Panth_AdvancedSEO; package mage2kishan/module-advanced-seo; namespace Panth\AdvancedSEO.
• Service contracts: Api\MetaResolverInterface (Model\Meta\Resolver), Api\CanonicalResolverInterface (Model\Canonical\Resolver), Api\RuleEvaluatorInterface (Model\Rule\Engine), Api\SeoScorerInterface (Model\Score\Scorer), Api\CustomCanonicalRepositoryInterface.
• Extension points via di.xml: add tokens to the tokens argument of Model\Meta\TokenRegistry (implement Model\Meta\Token\TokenInterface); add score checks and weights to the checks and weights arguments of Model\Score\Scorer (implement Model\Score\CheckInterface).
• Main plugins: Magento\Framework\View\Page\Config (pagination canonical, head meta, pagination meta, search result meta), View\Result\Layout::renderResult (rel prev/next, added after the product list is rendered), Page\Title (strip prefix/suffix), product/category/CMS metadata plugins, Magento\Catalog\Helper\Category and Helper\Product (native canonical suppression), CategoryUrlPathGenerator (short category URL), Catalog\Model\Product (visibility URL rewrites), frontend Response\Http (canonical Link header, Last-Modified) and View\Result\Page (SEO toolbar), plus admin form data provider and save plugins.
• Observers: product/category/CMS save, delete and category move (cache invalidation, reindex, scoring, cleanup), catalog_product_save_before (URL key), and Indexing API submission on save.
• Indexer panth_seo_resolved_meta; queue topic panth_seo.score_entity, consumer panth_seo.score_entity.consumer.
• Routes: admin panth_seo, frontend panth_seo.
• ACL: Panth_AdvancedSEO::manage with ::dashboard, ::templates, ::rules, ::bulkeditor, ::feeds, ::custom_canonical, ::reports, ::missing_meta, ::crawl_audit, ::audit, ::config.
• Tables: panth_seo_template, panth_seo_override, panth_seo_resolved, panth_seo_rule, panth_seo_score, panth_seo_meta_embedding, panth_seo_duplicate, panth_seo_custom_canonical, panth_seo_crawl_result, panth_seo_related, panth_seo_feed_profile, panth_seo_feed_field. Also adds the layered_navigation_canonical column to catalog_eav_attribute.
• Log file: var/log/panth_seo.log.

Uninstallation

The module has no uninstall script. The panth_seo_* tables, the layered_navigation_canonical column in catalog_eav_attribute, the EAV attributes seo_name, meta_robots and layered_navigation_canonical, the panth_seo/* values in core_config_data and generated feed files under pub/media/panth_seo/ remain until removed manually. After removal, Magento's own defaults for catalog/seo/category_canonical_tag and catalog/seo/product_canonical_tag apply again.

Support

• Product page: Advanced SEO Extension for Magento 2
• Contact: kishansavaliya.com/contact
• Email: [email protected]
• Issues: GitHub issues

Documentation

See USER_GUIDE.md.

License

Commercial software license. See LICENSE.txt in this repository.

Changelog

See CHANGELOG.md.

Links

• Website: kishansavaliya.com
• All extensions catalogue: kishansavaliya.com/magento-extensions.html
• GitHub: mage2sk/module-advanced-seo
• Packagist: mage2kishan/module-advanced-seo