RankReady – AI & LLM SEO for ChatGPT, Perplexity & Google AI
| 开发者 |
adityaarsharma
posimyththemes sagarpatel124 sandip111 |
|---|---|
| 更新时间 | 2026年6月1日 19:17 |
| PHP版本: | 7.4 及以上 |
| WordPress版本: | 7.0 |
| 版权: | GPL-2.0-or-later |
| 版权网址: | 版权信息 |
详情介绍:
RankReady is a WordPress plugin built for the AI search layer — the answers ChatGPT, Perplexity, Claude, Gemini, and Google AI Overviews show before anyone reaches a blue link. Drop it in alongside your existing SEO plugin (Rank Math, Yoast, AIOSEO — any of them) and start showing up in AI answers and citations. No conflicts. No replacement. Zero frontend bloat.
Visit the official RankReady page →
Traditional SEO plugins optimize for Google's classic results. RankReady adds the layer above them: the AI SEO signals — llms.txt, FAQ schema, Markdown endpoints, AI crawler controls — that decide whether AI engines read and cite your content. This is generative engine optimization (GEO) and answer engine optimization (AEO) for WordPress, built to work with the SEO plugin you already use.
A quick walkthrough of the whole plugin.
https://www.youtube.com/watch?v=JA-rEwMbqNo
Built by POSIMYTH Inc. — the team behind The Plus Addons for Elementor, NexterWP, and UiChemy.
Coexists with your SEO plugin — zero frontend impact
Install RankReady, optionally pick an LLM provider (OpenAI, Anthropic, Gemini, or DeepSeek) for the AI Summary and FAQ generators, and you're set. It auto-detects your active SEO plugin and never emits duplicate schema — your existing Yoast or Rank Math setup keeps working exactly as before. All AI generation runs in the WordPress admin, so there are no API calls on page load, no third-party scripts, and no extra requests for your visitors. Core Web Vitals are unaffected.
llms.txt — the AI-native sitemap
RankReady serves the llmstxt.org standard at
/llms.txt (a curated index of your best content) and /llms-full.txt (the full content concatenated as Markdown). AI crawlers read these files first to understand your site. Configurable post types, max post count, category and tag exclusions, and a per-domain brand identity (site name, summary, about section) you control from the AI Crawlers tab. Multilingual sites get hreflang Link headers when WPML, Polylang, TranslatePress, Weglot, or GTranslate is detected.
AI Summary generator with Speakable schema
Generate "Key Takeaways" for any post or page via your chosen LLM (OpenAI, Anthropic Claude, Google Gemini, or DeepSeek). The summary injects above your content as a styled block with Speakable schema — the JSON-LD that voice assistants read aloud. Use the Regenerate button in the post editor, the Gutenberg block, or the Elementor widget. Unlimited manual generations.
FAQ schema generator with DataForSEO
A strong signal for AI Overviews. RankReady can query DataForSEO for the real "People Also Ask" questions ranking for your post's focus keyword, then has your chosen LLM write the answers. Output is FAQPage JSON-LD — the structured data Google AI Overviews and Perplexity frequently cite over plain article text. Don't use DataForSEO? Type your own questions and let the LLM answer them. Unlimited manual generations. Setup guide in the FAQ section below.
Author Box with basic E-E-A-T schema
E-E-A-T (Experience, Expertise, Authoritativeness, Trustworthiness) is what AI models use to decide which sources to trust. RankReady ships a basic Author Box — name, job title, employer, bio, headshot, and basic sameAs links — with Article, Speakable, and FAQPage schema. It auto-detects Rank Math, Yoast, and AIOSEO and skips duplicate output. Display it anywhere via the Gutenberg block or the Elementor widget.
Markdown endpoints for AI agents
Every published post and page is served as clean Markdown at /post-slug.md with YAML frontmatter (title, author, dates). AI agents — Claude Desktop, Cursor, ChatGPT, custom clients — read Markdown faster than HTML. Content negotiation via Accept: text/markdown lets crawlers fetch the format they prefer with no URL changes.
MCP via the WordPress Abilities API
RankReady registers read-only abilities with the Model Context Protocol through the WordPress Abilities API (wp_register_ability()). On WordPress 7.0 these are surfaced by the official MCP Adapter — no bundled MCP server, no extra service to run.
Insights — AI referrals, bot activity, freshness
The Insights tab gives you real, server-side analytics with no third-party scripts:
- Training & Citation Bots — Which AI crawlers fetched which pages (GPTBot, ClaudeBot, PerplexityBot, OAI-SearchBot, Google-Extended, and more). Each citation-bot hit is a live AI answer that retrieved your page.
- Real AI Referrals — Humans clicking through from chatgpt.com, perplexity.ai, claude.ai, gemini.google.com, and copilot.microsoft.com, tracked via the HTTP Referer header.
- Content Freshness scanner — Buckets posts into Stale, Going stale, and Fresh, with one-click
dateModifiedrefresh to signal recency to AI crawlers.
robots.txt — both the WordPress virtual robots.txt and a physical ABSPATH/robots.txt if another plugin intercepts the URL. Plus Content Signals (ai-train, search, ai-input directives per contentsignals.org).
Cache compatibility + Diagnostics
RankReady persists cache-bypass entries to each cache plugin's stored configuration so server-level caches honour the bypass before PHP runs — tested with WP Rocket, LiteSpeed Cache, W3 Total Cache, WP Super Cache, WP Fastest Cache, Breeze, SG Optimizer, Hummingbird, Cache Enabler, Comet Cache, Swift Performance, NitroPack, Perfmatters, Cloudflare APO, Pantheon, Kinsta, and WP Engine. The Diagnostics card runs live endpoint probes, detects active SEO plugins, checks rewrite rules and REST routes, scans for cache conflicts, and gives you a one-click plain-text report for support — every failure ships with a one-line fix.
Coming soon
A companion add-on (in development) will layer advanced AI SEO automation on top of the free build:
- Auto-generate AI Summaries and FAQs on publish
- Bulk-generate summaries and FAQs across your whole library
- HowTo and ItemList JSON-LD
- Advanced Person / E-E-A-T schema (credentials, education, certifications, memberships, awards, Wikidata / ORCID / Scholar / LinkedIn, editorial & fact-check policies, Author Trust Panel)
- Custom post type support beyond posts and pages
- Headless / WPGraphQL API for decoupled front ends
安装:
Easy install (recommended)
- In WordPress admin, go to Plugins → Add New.
- Search for "RankReady".
- Click Install Now, then Activate.
- Visit RankReady in the admin menu.
- Add your AI provider API key (OpenAI, Anthropic, Gemini, or DeepSeek) in the Settings tab.
- Optionally enable llms.txt, Markdown endpoints, and AI crawler controls in the AI Crawlers tab.
- Download the plugin zip from WordPress.org.
- Go to Plugins → Add New → Upload Plugin and select the zip.
- Activate, then follow steps 4 to 6 above.
- Visit your site at
/llms.txtto confirm the llms.txt file is being served. - Open any post and use the AI Summary meta box to generate your first summary.
- Add the RankReady Author Box Gutenberg block (or Elementor widget) to a post to display the author bio.
屏幕截图:
常见问题:
Will RankReady conflict with Rank Math, Yoast, or AIOSEO?
No. RankReady is designed to work alongside Rank Math, Yoast, All in One SEO, SEOPress, SEO Framework, and Slim SEO. Before injecting any schema, it checks if another schema-generating plugin is active. If yes, it skips its own output or merges fields into the existing schema graph via documented filters. Verifiable with Google's Rich Results Test — no duplicate Article, Person, or FAQPage nodes.
How does RankReady actually work?
Three layers: (1) it serves discovery files (/llms.txt, /llms-full.txt, /post-slug.md) that AI crawlers read to find your content faster; (2) it adds AI-specific schema (FAQPage, Speakable, Article JSON-LD) that AI engines cite; (3) it gives you controls over which AI bots see your content, plus Insights analytics on which ones already do. It also registers read-only MCP abilities through the WordPress Abilities API so AI agents can discover your content.
Will this slow down my site?
No. All AI generation happens in the WordPress admin (not on page load). Schema and discovery headers add a few hundred bytes per page. The llms.txt and robots.txt files are cached via a 10-minute transient with stale-while-revalidate. Page Speed Insights and Core Web Vitals: unaffected.
Do I need an AI provider API key?
Only if you want to use the AI Summary or FAQ generators. The llms.txt generator, Markdown endpoints, AI crawler controls, Article schema, Author Box, AI referral tracking, content freshness scanner, and MCP abilities all work without any API key.
Are there usage limits or monthly caps?
No caps. Manual AI Summary generation and FAQ generation are unlimited. You pay only your own LLM API usage (typically $0.001 to $0.01 per generation). All features in the free build work with no limits.
Which AI provider should I pick?
All four work great. Practical guidance:
- OpenAI (
gpt-4o-mini,gpt-5) — Best all-rounder, widest model choice, predictable output. Recommended default. Pay-as-you-go at platform.openai.com. - Anthropic Claude (
claude-sonnet-4,claude-opus-4) — Strongest at long-form summaries and faithful citations. Recommended for long posts (3,000+ words). Console at console.anthropic.com. - Google Gemini (
gemini-2.5-flash,gemini-2.5-pro) — Generous free tier (up to 1,500 requests/day on Flash). Recommended to test before paying. Get a key at aistudio.google.com. - DeepSeek (
deepseek-v4-flash,deepseek-v4-pro) — Cheapest paid option, open-source models. Recommended for high-volume sites. Sign up at platform.deepseek.com.
How do I set up DataForSEO for the FAQ Generator?
The FAQ Generator uses DataForSEO to discover real "People Also Ask" questions for each post's focus keyword. Setup walkthrough:
- Create a DataForSEO account at dataforseo.com/register. The first $1 of credit is free for new sign-ups — enough for ~200 keyword lookups.
- After confirming your email, log in to the DataForSEO dashboard.
- Go to Settings → API Access. Copy your Login (your account email) and Password (an API password DataForSEO generates separately from your dashboard login).
- In WordPress, go to RankReady → Settings. Scroll to the DataForSEO card.
- Paste the Login and Password fields. Click Verify credentials — RankReady performs a live test query and shows your remaining account balance.
- Open any post, scroll to the RankReady FAQ meta box, enter a focus keyword, and click Generate questions. DataForSEO returns 5 to 10 real Google "People Also Ask" questions for that keyword.
- Pick which questions to keep, then click Generate answers to have your chosen LLM write the answers. Final FAQPage JSON-LD is auto-injected into the post.
What is an "llms.txt" file?
llms.txt is an emerging standard (llmstxt.org) that lets AI models like ChatGPT, Perplexity, and Claude understand your site's structure faster. Think of it as an "AI sitemap" — a curated index of your most important content optimized for LLM consumption. RankReady generates both /llms.txt (index) and /llms-full.txt (full content) automatically.
What is MCP and how does RankReady use it?
Model Context Protocol (MCP) is an open standard for letting AI agents discover and read your site's structured content. RankReady registers read-only abilities (read posts, list authors, fetch FAQs, query categories) through the WordPress Abilities API. On WordPress 7.0 these are surfaced by the official MCP Adapter — there is no bundled MCP server to run.
How does the freshness scanner work?
In Insights → Content Fresh, click Scan Content Freshness. RankReady reads every post's post_modified date and buckets them into Stale (60+ days), Going stale (30-59 days), and Fresh (under 30 days). Tick the boxes next to stale posts, click Refresh dateModified, and RankReady updates the modified timestamp without changing your content. This signals recency to AI crawlers on their next visit.
Does this work with my caching plugin or Cloudflare?
Yes. RankReady is tested with WP Rocket, LiteSpeed Cache, W3 Total Cache, WP Super Cache, WP Fastest Cache, Breeze, SG Optimizer, Hummingbird, Comet Cache, Cache Enabler, Swift Performance, NitroPack, Perfmatters, Cloudflare APO, Pantheon, Kinsta, and WP Engine. The plugin persists cache-bypass entries to each cache plugin's stored configuration so server-level caches honour the bypass before PHP runs. If your CDN still caches stale /llms.txt, copy the .htaccess or nginx snippet from Settings → Diagnostics → Server bypass snippets and add it to your server config.
Why is my Cloudflare edge serving a stale /llms.txt?
If you're on Cloudflare (especially with APO or a "Cache Everything" page rule), the edge can hold /llms.txt for up to 30 days. RankReady v1.0.0 sets s-maxage=600, CDN-Cache-Control, and Cloudflare-CDN-Cache-Control headers so the edge respects a 10-minute TTL. After updating, purge /llms.txt once in Cloudflare → Caching → Custom Purge by URL to flush any previously-cached version. Future updates auto-purge.
Where is my data stored?
Everything stays on your own WordPress site. Your API keys, DataForSEO credentials, generated summaries, FAQs, and author profiles all live in your own wp_options and wp_postmeta tables. POSIMYTH does not see, collect, or transmit any of your data.
How do I check if it is actually working?
Open Settings → Diagnostics and click Run Diagnostics. RankReady performs 26 live probes — fetches /llms.txt, /llms-full.txt, /.well-known/mcp.json, every Markdown route, detects active SEO plugins, checks rewrite rules, tests REST routes, scans for cache-plugin conflicts, and inspects edge cache headers. Every failure ships with a one-line fix. Click Copy Diagnostic Report for a full plain-text bundle you can paste into support requests.
How do I uninstall it cleanly?
By default, RankReady preserves your data on uninstall — your settings, API keys, summaries, FAQ data, and author profiles all survive. If you want a complete wipe, enable the "Delete all data on uninstall" toggle in the Advanced → Tools tab before uninstalling.
Is the source code available?
Yes. RankReady is open source under GPL-2.0-or-later. The complete source ships in the plugin zip on WordPress.org, and product info lives at posimyth.com.
更新日志:
1.1.0 — 2026-06-01
Minor release — blocks/widgets polish, multi-provider model auto-migration, leaner build.
- Consistent block/widget names across Gutenberg and Elementor — "AI Summary — RankReady", "FAQ — RankReady", "Author Box — RankReady". New dedicated "RankReady" group in both the block inserter and the Elementor widgets panel.
- Smart Generate/Regenerate button — shows "Generate" the first time and "Regenerate" once content exists, on any post type. Moved into the Elementor control panel as a full-width button. FAQ Elementor widget gains the Generate FAQ button and a Focus Keyword field (now matches the block).
- Inline model-migration notice — when a model you had selected is retired and auto-migrated to the current same-tier model, the notice now appears in the Settings tab body (next to where you pick the model) instead of as a global admin notice on every RankReady page.
- All AI providers detected everywhere — OpenAI, Claude, Gemini, and DeepSeek keys now unlock generation across blocks, widgets, REST, bulk gates, diagnostics, and the admin "API key" pill (previously only OpenAI was detected).
- Current model lists per provider, with automatic migration of retired model IDs to the current same-tier model.
- Clear, actionable error messages (missing key, out of credits, retired model, not enough content, and more).
- Fixed a false "Brand Identity not filled" warning in Diagnostics.
- Fixed AI-crawler checkbox alignment in robots.txt settings.
- Onboarding now also guides existing sites once after an update — shown a single time, never loops.
- Markdown
.mdendpoints send proper public cache headers for CDN compatibility. - Password-protected, private, and draft content stays excluded from every AI-facing surface.
- Lighter front end — stylesheet loads only on pages that actually use a RankReady block or widget, with zero front-end JavaScript and fully namespaced markup, so it never conflicts with other plugins or page builders.
- No data loss on update — existing summaries, FAQs, author profiles, settings, and API keys are untouched.
- Homepage Markdown URL fixed. Sites using a static front page (Reading Settings → "A static page") were emitting
<link rel="alternate" type="text/markdown" href="https://example.com.md" />— that's a different domain entirely (.mdis the Moldova country TLD). Now correctly emits/index.mdand resolves it back to the front-page post. - AI Summary settings now save. The auto-generate, auto-display, position, label, heading-tag, custom-prompt, and post-types fields silently dropped on save because they were registered against the wrong settings group. Fixed at the source —
options.phpnow whitelists every Content AI field correctly. - WP-Cron diagnostic accepts external cron. No more false warnings on RunCloud / Kinsta / WP Engine / Pantheon (sites with
DISABLE_WP_CRON=truerunning real system cron). The probe now measures whether scheduled events actually fire on time, regardless of which mechanism owns them. Up to 20 minutes overdue = healthy. - Complex-cache content-negotiation fix. The fix for sites running behind Cloudflare, Varnish, Fastly, or any CDN that doesn't vary cache key by
Acceptheader.Cloudflare-CDN-Cache-Control: no-storeauto-emitted on every markdown-eligible page request that specifiesAccept: text/markdown. Markdown requests now bypass cache, reach origin, return correct content-type. - Cloudflare APO auto-detect. When APO is detected and Markdown endpoints are enabled, RankReady surfaces a dismissible admin notice with the exact Cloudflare Cache Rule snippet to paste — applying it makes content-negotiated Markdown work without any cache buster.
- Cache headers audited to RFC 9110/9111 + enterprise CDN consensus. Removed
Pragma(response-direction RFC violation). AddedExpires: 0for HTTP/1.0 proxies. AddedAkamai-Cache-Controlalongside legacyEdge-Control. Standardised to Title-Case everywhere. AddedVary: Accept-Encodingto everyVaryheader. Added ETag +If-None-Match→ 304 on/llms.txtand/.well-known/mcp.json(60–90% bandwidth saving on revalidation). Splitmax-agevss-maxageso browsers re-check often (cheap with ETag) while CDN edges cache longer. Added open CORS on AI-readable endpoints so cross-origin AI agents can fetch. - UX polish. Removed the "Enable [feature]" intermediate button across 7 togglable cards (LLMs.txt, Markdown, robots.txt, Content Signals, WebMCP, Author Box, Schema Types). Users now tick the master toggle and hit Save — single conventional WordPress settings flow. Form rows breathe (16px row padding, 22px conditional-section margin, row dividers).
- Pro Extension API —
rnrd_is_pro()is now filterable viaapply_filters( 'rnrd_is_pro', false ). Newdo_action( 'rnrd_loaded' )fires after every core class init. Companion Pro addons attach via standard WordPress hooks — no mu-plugin trick required. - Build hygiene. Release zip now strips
CHANGELOG.md,README.md,docs/,.wp-org-prefixand every VCS/IDE/OS dotfile viaplugin/build-release.sh. Smaller zip, nothing dev-only reaches WP.org review.
- Security: private and password-protected content is never exposed on any AI surface. Markdown
.mdendpoints, llms.txt / llms-full.txt, the MCP abilities, the headless REST routes, the core/wp/v2meta fields, and the front-end JSON-LD schema injectors all exclude non-published and password-protected posts. Queries usehas_password => falseand every single-post gate checkspost_password, mirroringvalidate_public_post(). Previously a password-protected post could leak its body via/slug.mdeven though its HTML showed the password form. - 3-step onboarding wizard. Step 1 captures Brand Identity (site name, one-line summary, about), Step 2 applies the recommended AI-visibility stack (llms.txt, Markdown, robots, Content Signals, auto-summaries, FAQs) with one click, Step 3 confirms and redirects to the Dashboard. Re-runnable from the Dashboard.
- All options save and persist correctly. Cross-form nulling between
LLMS_GROUP,CONTENT_GROUP,AUTHORITY_GROUP, andCRAWLERS_GROUPfully resolved; hidden inputs preserve unchecked toggles.rnrd_max_snippet_defaultnow defaults'on'and survives every save. - API keys encrypted at rest. OpenAI / DataForSEO / Cloudflare credentials are stored AES-256-GCM with an
rnrdenc:prefix; reads decrypt transparently, writes encrypt automatically. - Cloudflare cache control. Settings card detects Cloudflare and can add a cache-bypass rule for
.mdrequests so APO never serves cached HTML to agents. Email + API token only; zone auto-detected. - MCP via the WordPress Abilities API. RankReady registers its read-only abilities with
wp_register_ability(). On WordPress 7.0 these are core-native and surfaced by the official MCP Adapter — no bundled MCP server. - Lean Free build. Pro-only capabilities show a "Coming Soon" placeholder and ship no functional code in the Free zip: Headless Public API (CORS, revalidation, WPGraphQL), advanced Person/EEAT schema, Editorial & Fact-Check policy URLs, Author Trust Panel, HowTo & ItemList schema, auto-generation of summaries and FAQs, bulk tools, custom-post-type support beyond Posts and Pages, and white-label branding.
- Multilingual
.mdendpoints. WPML / Polylang per-language Markdown with hreflang alternates andVary: Accept-Language. - Branded Author Profile card on
profile.php/user-edit.phpwith a deep link to the E-E-A-T settings tab. - CSS isolation. Every WP-core override in
admin.cssis scoped under.rnrd-wrap; core form fields, buttons, and tables outside the RankReady page stay untouched.
- Homepage Markdown URL fixed. Sites using a static front page (Reading Settings → "A static page") were emitting
<link rel="alternate" type="text/markdown" href="https://example.com.md" />— that's a different domain entirely (.mdis the Moldova country TLD). Now correctly emits/index.mdand resolves it back to the front-page post. - AI Summary settings now save. The auto-generate, auto-display, position, label, heading-tag, custom-prompt, and post-types fields silently dropped on save because they were registered against the wrong settings group. Fixed at the source —
options.phpnow whitelists every Content AI field correctly. - WP-Cron diagnostic accepts external cron. No more false warnings on RunCloud / Kinsta / WP Engine / Pantheon (sites with
DISABLE_WP_CRON=truerunning real system cron). The probe now measures whether scheduled events actually fire on time, regardless of which mechanism owns them. Up to 20 minutes overdue = healthy. - Complex-cache content-negotiation fix. The fix for sites running behind Cloudflare, Varnish, Fastly, or any CDN that doesn't vary cache key by
Acceptheader.Cloudflare-CDN-Cache-Control: no-storeauto-emitted on every markdown-eligible page request that specifiesAccept: text/markdown. Markdown requests now bypass cache, reach origin, return correct content-type. - Cloudflare APO auto-detect. When APO is detected and Markdown endpoints are enabled, RankReady surfaces a dismissible admin notice with the exact Cloudflare Cache Rule snippet to paste — applying it makes content-negotiated Markdown work without any cache buster.
- Cache headers audited to RFC 9110/9111 + enterprise CDN consensus. Removed
Pragma(response-direction RFC violation). AddedExpires: 0for HTTP/1.0 proxies. AddedAkamai-Cache-Controlalongside legacyEdge-Control. Standardised to Title-Case everywhere. AddedVary: Accept-Encodingto everyVaryheader. Added ETag +If-None-Match→ 304 on/llms.txtand/.well-known/mcp.json(60–90% bandwidth saving on revalidation). Splitmax-agevss-maxageso browsers re-check often (cheap with ETag) while CDN edges cache longer. Added open CORS on AI-readable endpoints so cross-origin AI agents can fetch. - UX polish. Removed the "Enable [feature]" intermediate button across 7 togglable cards (LLMs.txt, Markdown, robots.txt, Content Signals, WebMCP, Author Box, Schema Types). Users now tick the master toggle and hit Save — single conventional WordPress settings flow. Form rows breathe (16px row padding, 22px conditional-section margin, row dividers).
- Pro Extension API —
rnrd_is_pro()is now filterable viaapply_filters( 'rnrd_is_pro', false ). Newdo_action( 'rnrd_loaded' )fires after every core class init. Companion Pro addons attach via standard WordPress hooks — no mu-plugin trick required. - Build hygiene. Release zip now strips
CHANGELOG.md,README.md,docs/,.wp-org-prefixand every VCS/IDE/OS dotfile viaplugin/build-release.sh. Smaller zip, nothing dev-only reaches WP.org review.
- Unlimited AI Summaries — generate "Key Takeaways" for every post via OpenAI, Anthropic, Gemini, or DeepSeek with zero per-month limit on manual generation.
- FAQ schema generator — discover real user questions via DataForSEO, answer via your chosen LLM, output FAQPage JSON-LD that AI Overviews preferentially cite.
- llms.txt + Markdown endpoints — serve the llmstxt.org standard at
/llms.txt,/llms-full.txt, and every post as Markdown at/post-slug.mdfor AI agents. - 31+ AI crawler controls — granular allow/block for GPTBot, ClaudeBot, PerplexityBot, Google-Extended, Applebot-Extended, and 26 more, auto-synced to
robots.txt. - E-E-A-T schema + Author Box — Article, Speakable, and Author JSON-LD signals out of the box, designed to coexist with Rank Math, Yoast, AIOSEO without duplicate output.
- Content Freshness scanner — bucket posts into Stale / Going Stale / Fresh, bulk-refresh
dateModifiedto signal recency to AI crawlers. - Insights dashboard — Training Bots, Citation Bots, Real AI Referrals, and Content Fresh analytics with demo-data preview.
- Cache compatibility — first-class support for WP Rocket, LiteSpeed, W3TC, WP Super Cache, WP Fastest Cache, Breeze, SG Optimizer, Hummingbird, Cache Enabler, Comet Cache, Swift Performance, NitroPack, Perfmatters, Cloudflare APO, Pantheon, Kinsta, WP Engine.
- Multilingual llms.txt — auto-detects WPML, Polylang, TranslatePress, Weglot, GTranslate; emits hreflang Link headers.
- Diagnostics suite — 26 live endpoint probes, plugin-conflict detection, and copy-ready Apache/Nginx server-bypass snippets.