Title: FoundByAI — LLMs.txt & AI Bot Tracking Author: thenooch Published: July 10, 2026 Last modified: October 2, 2026 --- Search plugins ![](https://ps.w.org/seo-for-ai-bots/assets/banner-772x250.png?rev=3603410) ![](https://ps.w.org/seo-for-ai-bots/assets/icon-256x256.png?rev=3603398) # FoundByAI — LLMs.txt & AI Bot Tracking By [thenooch](https://profiles.wordpress.org/thenooch/) [Download](https://downloads.wordpress.org/plugin/seo-for-ai-bots.1.3.5.zip) * [Details](https://pcd.wordpress.org/plugins/seo-for-ai-bots/#description) * [Reviews](https://pcd.wordpress.org/plugins/seo-for-ai-bots/#reviews) * [Installation](https://pcd.wordpress.org/plugins/seo-for-ai-bots/#installation) * [Development](https://pcd.wordpress.org/plugins/seo-for-ai-bots/#developers) [Support](https://wordpress.org/support/plugin/seo-for-ai-bots/) ## Description AI tools like ChatGPT, Claude, and Perplexity now send real traffic to the sites they cite — but most WordPress pages are invisible to them. FoundByAI shows you exactly what’s holding each page back, then fixes the plumbing for you: it audits every page for AI readiness, serves clean Markdown to AI crawlers, and builds your llms.txt automatically. Built specifically for AI visibility and Generative Engine Optimization (GEO) — also called AI search optimization — not bolted onto a legacy SEO plugin. It logs visits from the AI crawlers listed in the Bots tab, runs a per-page AI audit with concrete fixes, and works with or without Yoast, Rank Math, and AIOSEO — reading their scores when present, and writing your titles and meta descriptions itself when no SEO plugin is active. #### Who is this for? * ✅ **Bloggers & publishers** who want to get quoted when people ask an AI a question * ✅ **Store owners** (WooCommerce) who want products discoverable by AI shopping assistants * ✅ **Marketers & agencies** managing AI visibility across client sites No code required, and no need to replace the SEO plugin you already use. #### What FoundByAI does for you **Content Serving** * .md URL Endpoints: Visit `/post-slug.md` to download any post as clean Markdown * Custom Post Type Support: Serve Markdown for Posts, Pages, or custom post types * WooCommerce Integration: Auto-detects WooCommerce and displays dedicated settings for product discovery * YAML Frontmatter: Each Markdown file includes metadata in YAML format **AI Discovery (llms.txt)** * Dual Endpoints: Generate llms.txt at both `/.well- known/llms.txt` and `/llms.txt` * Link Descriptions: Every entry carries a one-line description, taken from your SEO meta description, excerpt, or content — so AI can tell what a page covers before fetching it * Markdown Targets: Entries link to the clean `.md` version of each page, not the HTML page wrapped in navigation and scripts* Full-Content File: `/llms-full.txt` embeds the complete Markdown of every listed page, per the llmstxt.org spec * Auto-Detection: Auto-discover and prioritize by SEO plugin scores * 3-Tier Prioritization: SEO score > post date > manual pinning rules * Popular With AI: Highlights the pages AI crawlers actually read most, based on your own bot traffic * Always Current: Rebuilt whenever you publish or edit, not on a fixed schedule * Conflict Detection: Tells you when another plugin’s llms. txt file is overriding yours, and helps you clear it * Smart Filtering: Auto-exclude <200 word posts, password-protected, private, noindex content * 1-Hour Cache: Intelligent caching with smart invalidation * Dynamic Pin Rules: Boost posts by tag, category, or manual selection **Dynamic Bot Management** * Curated Bot List: the main AI training, search and user-triggered crawlers, editable in the Bots tab * Enable/Disable Controls: Manage which bots to track and serve * Extensibility Filter: Register custom bots with ` sfab_register_bots` **SEO Plugin Integration** * Auto-Detect: Yoast SEO, Rank Math, All In One SEO * Score Reading: Use SEO scores for llms.txt prioritization * Noindex Respect: Respect SEO plugin’s noindex settings * Manual Override: Force specific plugin for edge cases **Advanced Controls** * Per-Post Discovery: Include/exclude individual posts from llms.txt * Word Count Threshold: Filter posts by minimum word count * ACF Field Support (Optional): Include custom fields in Markdown frontmatter * Bot Tracking: Monitor which AI crawlers visit your content * Logs Dashboard: View and analyze bot visit history **Admin UI** * Configuration Tabs: Markdown Settings, llms.txt, robots.txt, Sitemap, Site Info, Plugin Settings, API Keys, Connect AI Tools * llms.txt Preview: See generated discovery file in admin * Pin Rules Manager: Add/remove tag, category, and post- specific boosts * Responsive Design: Works on desktop and mobile #### How It Works 1. **Markdown endpoints** are advertised with a `rel="alternate"` link and available at `.md` URLs 2. **Plugin serves** clean Markdown + YAML frontmatter 3. **Discovery file** (llms.txt) lists your best posts in priority order (optional) 4. **Analytics** track which AI crawlers visit and how they engage 5. **SEO stays intact** – no interference with your SEO plugin #### Strategic Positioning This plugin is a **specialist tool**, not a competitor to SEO plugins: * ✅ Owns: Markdown serving, AI crawler discovery, llms.txt generation, AI bot analytics * ✅ Complements: Yoast, Rank Math, All In One SEO (reads their scores, respects their rules) * ✅ Smart robots.txt: Auto-injects llms.txt rules when safe, shows manual instructions otherwise * ✅ Stays out of: XML sitemaps, core SEO meta tags #### Compatibility ✅ Yoast SEO, Rank Math, All In One SEO (full integration) ✅ WooCommerce (dedicated product discovery settings) ✅ Advanced Custom Fields (optional ACF field support) ✅ All custom post types ✅ WordPress 5.0+ and PHP 7.4+ #### Test It After activation, test your setup: 1. Visit SEO for AI Bots > Settings 2. View any tab to verify configuration 3. Visit your site’s `/llms.txt` endpoint 4. Visit any post with `.md` extension: `yoursite.com/your-post.md` 5. View bot visits in SEO for AI Bots > Bot Visits Log ### External Services This plugin optionally connects to the following external services. All connections are initiated by user action or explicitly opt-in — the core plugin functions entirely offline. #### 1. FoundByAI Proxy (seoforaibots.com) * **What it does:** Powers the WAF/firewall bot-access check (verifies AI crawlers can reach your site from an outside vantage point), and receives anonymized usage telemetry if you opt in. It does NOT run the AI Search Simulator — those calls go directly from your site to the AI providers using your own keys (see below). * **Data sent:** Your site URL (for the firewall/bot-access check); an anonymized site hash and plugin version (telemetry only). Telemetry is never sent unless you explicitly enable it. * **When:** The firewall check runs only when you click “Check”. Telemetry is weekly and only if you opt in under Settings > Privacy (off by default). The plugin does not fetch configuration or contact this server on its own. * **Terms of Service:** https://seoforaibots.com/terms-of-service/ * **Privacy Policy:** https://seoforaibots.com/privacy-policy/ #### 2. OpenRouter (openrouter.ai) — bring-your-own-key * **What it does:** Runs the AI Search Simulator’s query generation and the ChatGPT/ Claude / Gemini answer previews. Requests are sent directly from your site with your own OpenRouter API key — FoundByAI never sees your key and never proxies these calls. * **Data sent:** Your OpenRouter API key (as the request credential), plus the generated query and a short excerpt of the post being tested. * **When:** Only when you have entered an OpenRouter key under Configuration > AI Keys and click “Run” in the AI Search Simulator. * **Terms of Service:** https://openrouter.ai/terms * **Privacy Policy:** https://openrouter.ai/privacy #### 3. Perplexity (api.perplexity.ai) — bring-your-own-key, optional * **What it does:** Runs the AI Search Simulator’s live web-search visibility check. Requests are sent directly from your site with your own Perplexity API key. * **Data sent:** Your Perplexity API key and the search query being tested. * **When:** Only when you have entered a Perplexity key under Configuration > AI Keys and run a live search. * **Terms of Service:** https://www.perplexity.ai/hub/legal/terms-of-service * **Privacy Policy:** https://www.perplexity.ai/hub/legal/privacy-policy #### 4. Google Gemini API (generativelanguage.googleapis.com) — bring-your-own-key, optional * **What it does:** Runs the optional direct Gemini visibility simulation. Requests are sent directly from your site with your own Google AI Studio API key. * **Data sent:** Your Gemini API key, plus the query and a short excerpt of the post being tested. * **When:** Only when you have entered a Gemini key under Configuration > AI Keys and run the Gemini simulation. * **Terms of Service:** https://ai.google.dev/gemini-api/terms * **Privacy Policy:** https://policies.google.com/privacy #### 5. Google Search Console API (googleapis.com) — opt-in * **What it does:** Retrieves your site’s search performance data to power the AI Coverage Gap dashboard (shows pages with Google traffic but no AI bot visits). * **Data sent:** OAuth access token for authentication; your site’s GSC property URL to filter results. * **When:** Only when you connect GSC via the AI Keys tab (one-click OAuth flow) and visit the AI Coverage Gap dashboard. No data is sent until you initiate the connection. * **Terms of Service:** https://policies.google.com/terms * **Privacy Policy:** https://policies.google.com/privacy #### 6. Wikipedia API (en.wikipedia.org) — on-demand * **What it does:** Checks whether your brand has a Wikipedia article, shown as an advisory in the AI Readiness dashboard. * **Data sent:** Your brand name (from plugin settings). * **When:** Only when you view the AI Readiness dashboard with a brand name configured. Results are cached for 24 hours. * **Terms of Use:** https://foundation.wikimedia.org/wiki/Policy:Terms_of_Use * **Privacy Policy:** https://foundation.wikimedia.org/wiki/Policy:Privacy_policy #### 7. DataForSEO (api.dataforseo.com) — bring-your-own-key, optional * **What it does:** Powers the AI Search Simulator’s optional “Real Citations” mode. Instead of simulating an answer, FoundByAI asks DataForSEO to query the real ChatGPT, Claude, Gemini, and Perplexity and return their actual answers and cited sources. Requests are sent directly from your site with your own DataForSEO API credentials and billed to your DataForSEO account. * **Data sent:** Your DataForSEO API login and password (as the request credential) and the search query being tested. * **When:** Only when you have entered DataForSEO credentials under Configuration > AI Keys, enabled “Real Citations” mode (off by default), and run a check. * **Terms of Service:** https://dataforseo.com/terms-of-use * **Privacy Policy:** https://dataforseo.com/privacy-policy ### Support For issues, feature requests, or bug reports, please use the WordPress.org support forum or visit: https://seoforaibots.com ### Translations This plugin is ready for translation. Contribute at: https://translate.wordpress. org/projects/wp-plugins/seo-for-ai-bots/ ### Additional Resources * [Dries Buytaert’s “The Third Audience”](https://dri.es/the-third-audience) – The concept behind this plugin * [llms.txt Standard](https://llmstxt.com/) – AI-Friendly Discovery Format * [Content Negotiation (MDN)](https://developer.mozilla.org/en-US/docs/Web/HTTP/Content_negotiation) * [GPTBot Documentation](https://platform.openai.com/docs/gptbot) * [Markdown Guide](https://www.markdownguide.org/) * [Yoast SEO](https://yoast.com/wordpress/plugins/seo/) * [Rank Math](https://rankmath.com/) * [All In One SEO](https://aioseo.com/) ## Installation 1. Download the plugin 2. Go to Plugins > Add New > Upload Plugin 3. Select `seo-for-ai-bots.zip` and Install 4. Activate Plugin 5. Go to SEO for AI Bots > Settings to configure #### Manual Installation 1. Upload `seo-for-ai-bots` folder to `/wp-content/plugins/` 2. Activate through the Plugins menu 3. Configure in SEO for AI Bots ## FAQ ### What is llms.txt? llms.txt is a proposed index file (like robots.txt, but for language models) that lists the pages on your site that matter most, with links to clean Markdown versions. It is optional and unproven: Google Search ignores it, and no major AI search product has confirmed that it reads it. Some coding agents and developer tools do read it. FoundByAI generates and maintains one for you at no cost, but it is not the reason to install the plugin. ### Which AI bots does FoundByAI detect? It detects the main AI training, search and user-triggered crawlers from OpenAI, Anthropic, Perplexity, Apple and others. The full list is on the Bot Activity -> Bots tab, where you can enable or disable any bot and add your own patterns with the `sfab_register_bots` filter. ### Why can’t ChatGPT or Claude open my page when my browser can? Browsers are forgiving and AI fetchers are not. The usual causes are a firewall or CDN (Cloudflare, Sucuri, your host’s WAF) that challenges or blocks the crawlers’ user agents, a redirect loop, or a chain of hops such as `https -> http -> https` that a browser follows but an AI fetcher gives up on. The AI Access test on the Bot Activity -> Bots tab requests your site with each crawler’s user agent and follows redirects on your homepage and recent posts, then shows which ones are blocked and which pages redirect badly. The test runs from our server, not from OpenAI’s or Anthropic’s, so a WAF that filters by IP address can still treat the real crawlers differently. ### What is the difference between training, search and user-triggered bots? * **Training** crawlers (for example GPTBot and ClaudeBot) collect content in the background for model training. Blocking them does not stop AI assistants from reading your site today. * **Search** crawlers (for example OAI-SearchBot, Claude-SearchBot and PerplexityBot) build the index that AI search products cite from. * **User-triggered** fetchers (for example ChatGPT-User, Claude-User and Perplexity- User) load a page when a person asks an assistant about it. Blocking a search or user-triggered bot means that product cannot cite or open your page. ### Can I see Google-Extended or Applebot-Extended visits? No. Google-Extended and Applebot-Extended are robots.txt control tokens, not crawlers. They tell Google and Apple whether content the normal crawlers (Googlebot, Applebot) already fetched may be used for AI training. They never make a request of their own, so they cannot appear in a visit log. ### How do I enable llms.txt? llms.txt is enabled by default after activation. Visit SEO for AI Bots > Discovery to: * Preview your llms.txt file * Add pinning rules (boost posts by tag, category, or ID) * View posts that will be included ### Can I control which posts appear in llms.txt? Yes, three ways: 1. **Global threshold**: Settings > Content sets minimum word count (default 200) 2. **Per-post checkbox**: Edit any post to include/exclude from llms.txt 3. **Pinning rules**: Settings > Discovery to boost specific posts or categories ### What’s the difference between .md URLs and llms.txt? * **/.md URLs** (e.g., `/my-post.md`): Individual post downloads in Markdown format * **/llms.txt**: Directory file listing your best posts for AI crawler discovery ### My SEO plugin already generates an llms.txt. What happens? Only one llms.txt can win. Some plugins write a real file into your site’s root folder, and your web server hands that file to AI crawlers before WordPress even loads — so it silently overrides the one FoundByAI builds, whatever your settings say. FoundByAI detects this and tells you, rather than quietly competing. If you’d rather use FoundByAI’s version, turn the other plugin’s llms.txt setting off first (otherwise it will write the file again on its next scheduled run), then use the “Remove the file” button in the notice to clear the leftover. FoundByAI’s llms.txt differs in a few ways: it can list every page you choose rather than a fixed handful, it links to clean Markdown versions instead of your HTML pages, it gives each link a description, it rebuilds the moment you publish, and it can order entries by which pages AI crawlers actually read on your site. If you prefer the other plugin’s file, that’s fine too — just switch FoundByAI’s llms.txt off in Settings > Discovery. ### How does prioritization work? 3-tier system: 1. **SEO Score** (0-1000 points): From your SEO plugin (Yoast, Rank Math, etc.) 2. **Post Date** (0-100 points): Newer posts get higher score, decays over 1 year 3. **Manual Pins** (highest): Boost specific posts or categories ### Will it slow down my site? No. Processing only happens for: * Markdown requests (visitors requesting .md files or Markdown content type) * llms.txt generation (cached for 1 hour) * Regular page loads are unaffected ### Do I need an SEO plugin? No. The plugin works standalone. With an SEO plugin, it reads scores for smarter prioritization. Without one, it uses post dates and manual pinning rules. ### What if my SEO plugin isn’t detected? Go to Settings > Detection and manually select your plugin from the dropdown. This is rarely needed. ### Can I add custom AI bots? Yes, use the filter: `sfab_register_bots` Example: `php add_filter( 'sfab_register_bots', function( $bots ) { $bots[] = array(' pattern' => 'MyCustomBot', 'name' => 'My Custom AI Bot', 'enabled' => true, 'custom' => true, ); return $bots; } ); ### Does it work with ACF (Advanced Custom Fields)? Yes! If ACF is installed, go to Settings > Content to select which ACF fields to include in Markdown frontmatter. ### How do I test Markdown output? Method 1: Visit any post with `.md`: `yoursite.com/my-post.md` Method 2: Check llms.txt at: `yoursite.com/llms.txt` ### Can I customize llms.txt format? Use the filter: `sfab_llms_posts` to modify the posts list before generation. ### How many logs does it keep? Currently keeps last 500 bot visits with rolling expiration. Logs older than 14 days are auto-removed. ### Is there multisite support? Not in v2.0. Each site needs the plugin configured separately. ### Does this plugin collect any data? **Anonymous Telemetry (Opt-in, Disabled by Default)** The plugin includes optional anonymous telemetry. When you enable it in Settings > Privacy: * **Collected:** Anonymized site hash (not your URL), plugin version, active SEO plugin name, active FAQ plugin names, and aggregated bot visit counts (7-day totals) * **NOT collected:** Your site URL, page content, personal data, or user information * **Frequency:** Sent weekly to our analytics endpoint (Google Apps Script) * **Purpose:** Helps us understand which plugins to prioritize for compatibility * **Control:** Enable/disable anytime in Settings > Privacy **Bot Visit Logging (Local Only)** The Bot Activity feature stores visit data locally in your WordPress database: * **Stored:** Timestamp, page visited, bot name, user agent, IP address, request type * **Purpose:** Analytics to show which AI bots visit your content * **Retention:** Last 500 visits, auto-deleted after 14 days * **Location:** Your database only – never sent externally * **GDPR Note:** Bot IPs are typically datacenter IPs, not personal. If you prefer not to store IPs, you can disable bot tracking in Settings. ### Does the plugin connect to external services? The core plugin works entirely offline. Optional features that connect externally— see the “External Services” section below for full details on each service, what data is sent, and links to their terms and privacy policies. ## Reviews There are no reviews for this plugin. ## Contributors & Developers “FoundByAI — LLMs.txt & AI Bot Tracking” is open source software. The following people have contributed to this plugin. Contributors * [ thenooch ](https://profiles.wordpress.org/thenooch/) [Translate “FoundByAI — LLMs.txt & AI Bot Tracking” into your language.](https://translate.wordpress.org/projects/wp-plugins/seo-for-ai-bots) ### Interested in development? [Browse the code](https://plugins.trac.wordpress.org/browser/seo-for-ai-bots/), check out the [SVN repository](https://plugins.svn.wordpress.org/seo-for-ai-bots/), or subscribe to the [development log](https://plugins.trac.wordpress.org/log/seo-for-ai-bots/) by [RSS](https://plugins.trac.wordpress.org/log/seo-for-ai-bots/?limit=100&mode=stop_on_copy&format=rss). ## Changelog #### 1.3.5 – 2026-10-02 * Fix: the plugin no longer forces title-tag theme support. On older themes that print their own , every page had two title tags. #### 1.3.4 – 2026-09-29 * Fix: AI bot detection now names each agent separately: ClaudeBot (training), Claude-SearchBot (search) and Claude-User (a person asked Claude about your page); PerplexityBot and Perplexity-User; OAI-SearchBot and ChatGPT-User with the right type; Applebot. Previously Anthropic’s three agents were all logged as “Anthropic” and Perplexity’s two as “Perplexity Bot”, and a disabled entry let visits fall through to a generic one. If you had blocked “Anthropic”, the block still covers all Claude agents. * Removed: Claude-Web, OAI-Plug and Googlebot-Extended (retired or never real), plus Google-Extended and Applebot-Extended (robots.txt opt-out tokens, not crawlers, so they can never appear in a visit log), and a handful of bot names no vendor documents. You can still add any bot as a custom pattern. * Your recent visit log is re-labelled automatically on update. Older daily totals cannot be split and appear as “Anthropic (all agents, before 1.3.4)” and “Perplexity( all agents, before 1.3.4)”. * Removed the invented ai-description, ai-keywords, ai-optimized, ai-content-type and ai-content-format meta tags and the unused FAQ schema code. No AI system reads them. The rel=”alternate” Markdown link and Link header stay. * Scores no longer reward FAQPage, HowTo or Speakable schema, the llms.txt files, or an ai-sitemap rule. The schema score now looks at whether an Article is emitted, an author and a featured image. FAQPage and HowTo are labelled informational ( no rich result any more). * Fix: the robots.txt check no longer fails sites that correctly rely on “User- agent: _“. It now flags only a blocked search or user-triggered AI bot, and warns when a bot’s own group drops your “_” Disallow rules. * The AI Access check now probes Claude-SearchBot, Perplexity-User and Applebot, and no longer probes Google-Extended (no crawler sends that name) or python-requests/ python-httpx. The advice to allowlist those Python clients in your firewall is gone; it opened sites to generic scrapers. * Readme: a shorter, more accurate description and corrected claims. * Fix: Article schema is now only output for posts and pages. Before, any other enabled post type (a WooCommerce product, an event) was labelled an Article. Sites with an article-like custom type can opt it in with the `sfab_article_schema_post_types` filter. * Fix: sites updated from older versions no longer keep two dead scheduled events(` ai_markdown_server_daily_rescan_event` and `ai_markdown_server_refresh_analytics`) that fired on visitor page loads and did nothing. They are cleared once on update. * Removed unsourced statistics from three admin panels (“62% of AI citations are ghost citations”, “2.4x more brand mentions”, “research shows” claims about metadata). The advice stays; the numbers had no source we could point to. * New: A “Remove test visits” button on the Bot Activity Summary, Per-Post, Timeline and Recent tabs. Before 1.3.2 the AI Access check’s simulated bot requests were logged as real AI bot visits. The button finds an IP that, within a few seconds, sent five or more different AI bot names including at least two of the test’s own distinctive user-agent strings (real crawlers never do), removes only the visits from that IP, and rebuilds the daily stats. Visits from every other IP are left alone. It only appears when such visits exist. * Fix: the redirect check only tested the homepage and recent blog posts, so a site with no posts (a restaurant or brochure site) got just three homepage checks. It now uses recent pages when there are fewer than two posts. * The plugin’s name in the WordPress.org directory is now “FoundByAI — LLMs.txt& AI Bot Tracking”: brand first, with the search terms people actually use right after it. #### 1.3.2 – 2026-09-29 * Fix: the redirect-chain check added in 1.3.1 ran on our server but its results were not displayed anywhere in your admin. They now appear on the Bot Activity page (Bots tab) with the full hop-by-hop chain and fix steps, in the summary strip, and in the dashboard alert banner when a redirect problem is found. * Fix: the AI Access check’s own test requests were logged as real AI bot visits, so a new site showed “8 bot visits” that were all our test, and the Insights and Timeline tabs built recommendations from them. Test requests are now recognised and skipped. * Fix: Content Handlers described every “Extract” action as “structured Q&A”, including galleries and sliders. It now says “structured text”. * Fix: suggested alt text is cleaner. Filename suggestions no longer keep “+” characters or WordPress’s generated size suffix (e.g. “-300×225”). #### 1.3.1 – 2026-09-29 * New: The AI Access check now follows redirects hop by hop for your homepage and two most recent posts, including variants an AI fetcher might request (no trailing slash, http://, www). It flags redirect loops, chains longer than they should be, and any redirect that sends visitors from https to http://. Browsers and Google tolerate these; AI fetchers often give up, so a page can look fine to you and still fail to open in an AI assistant. The check runs from our server, not from OpenAI or Anthropic, so it approximates an external fetcher rather than reproducing one. * New: Claude-User is now included in the AI Access check. * Fix: a redirect loop is now reported as a redirect loop instead of “blocked by your firewall”. * New: The AI Checklist now flags insecure http:// images, scripts, stylesheets and embeds inside a post on an https site. It scans post content only, not theme files. * The plugin’s name in the WordPress.org directory is now “LLMs.txt & AI Bot Tracking for ChatGPT – FoundByAI”, so people searching for llms.txt and AI crawler tracking can find it. Your admin menu still reads FoundByAI. #### 1.3.0 – 2026-09-11 * New: Write your SEO title, meta description, canonical URL, and focus keyword directly in FoundByAI — no separate SEO plugin required. These fields appear on the post editor only when no dedicated SEO plugin (Yoast, Rank Math, All in One SEO) is active; if you have one, FoundByAI defers to it and stays out of the way. Includes a per-page nofollow toggle. * New: Optional “Real Citations” mode for the AI Search Simulator. Connect your own DataForSEO account and the simulator queries the real ChatGPT, Claude, Gemini, and Perplexity — showing each platform’s actual answer, the real sources it cited, and whether your site was cited, only mentioned (a “ghost citation”), or missed entirely. It runs on your DataForSEO credits at cost, is off by default, and the built-in simulation stays free. * Security: your DataForSEO credentials are encrypted at rest, like your other API keys, and are never rendered back into the settings form after saving. #### 1.2.0 – 2026-09-10 * FoundByAI is now 100% free. Every feature — AI Search Simulator, Competitor / Share-of-AI intelligence, per-page content handlers, the firewall bot-access check, and the SEO meta editor — is unlocked for everyone. There is no longer a Pro tier. * Bring your own key: the AI features run on your own provider API keys (OpenRouter, and optionally Perplexity and Google Gemini), calling each provider directly at cost with no markup. Add them under Configuration > AI Keys. * Security: your API keys are now encrypted at rest, stored non-autoloaded, and never rendered back into the settings form after saving. * Removed the Freemius licensing SDK. Updates continue to arrive normally through WordPress.org. * Added an optional, entirely voluntary “buy me a coffee” link for anyone who wants to support development. * Fixed: the robots.txt rules we suggest no longer include a “Sitemap:” line for/ ai-sitemap.xml. That directive asks search engines to index those URLs, but the Markdown endpoints it lists are intentionally noindex (so they can’t compete with your real pages) — which made site audits in Ahrefs and Semrush report “ Noindex page in sitemap”. AI crawlers find your Markdown through llms.txt and the alternate-link header, not through that directive, so nothing is lost. * Fixed: sites using WordPress’s virtual robots.txt now get an “Allow: /ai-sitemap. xml” rule. Previously only sites with a physical robots.txt file received it, so the AI Readiness check for that rule could never pass on a virtual robots. txt — even though the robots.txt screen reported it was being handled automatically. #### 1.1.11 – 2026-07-17 * Brand: The admin menu and screens now use the plugin’s FoundByAI name consistently( the “SEO for AI” menu label is now “FoundByAI”). * Fix: Markdown is now served only at its own .md URLs. Previously, when a bot or tool requested Markdown at one of your normal page URLs, a page cache could store that Markdown and then serve it to human visitors. Markdown is still advertised to AI via the Link header and llms.txt. * New: Every llms.txt entry now carries a one-line description, taken from your SEO meta description, your excerpt, or the content itself — so AI can tell what a page covers before fetching it. * New: Choose what llms.txt links to. Markdown stays the default; switch to the normal page globally under Discovery, or per post when editing, for shops, member areas, and pages whose value is in their structured data. * New: FoundByAI now detects when another plugin has written an llms.txt file into your site root — which your web server serves before WordPress loads, silently overriding ours — and offers to clear it once you’ve turned the other plugin’s setting off. * New: AI Readiness now includes a Ghost Citations advisory — practical guidance on getting AI assistants to name your brand in answers, not just cite your site anonymously. * Improved: Markdown endpoints now send X-Robots-Tag: noindex so search engines don’t index them as thin duplicates of your real pages, plus X-Content-Type-Options: nosniff. * Improved: Markdown files now include a canonical link to the page they mirror, so AI assistants can cite your real page rather than the .md file. * Fix: llms.txt no longer HTML-escapes its own text. Titles like “R&D” were reaching AI as “R&D”. * Fix: llms.txt links with query strings (used by some multilingual setups) were escaped for HTML and could arrive broken. * Pro: The AI Search Simulator’s live Perplexity search now uses your own Perplexity API key (bring-your-own-key). Add it under Configuration > API Keys to run live web searches directly at Perplexity’s cost, with no markup. The simulated ChatGPT, Claude, and Gemini previews continue to run automatically. #### 1.1.10 – 2026-06-22 * Privacy: Removed all automatic background calls to our server. The plugin no longer fetches remote configuration or bot definitions on a schedule — bot definitions ship with the plugin and update with each release. Nothing is sent off-site unless you explicitly opt in to telemetry. * Security: Custom JSON-LD entered per post is now decoded and re-encoded (sanitized) before it is stored, rather than saving the raw input. * Security: Speakable schema output no longer uses JSON flags that could weaken escaping inside the script tag. * Housekeeping: Removed the legacy daily config-refresh cron from existing installs. #### 1.1.9 – 2026-06-14 * Free: Content refactoring suggestions are now available to everyone — actionable structural tips to make each page more AI-friendly * Free: Per-post AI context and custom JSON-LD schema fields * Free: Per-post content handler controls (summarize / extract / remove how AI bots see specific content) * Free: WooCommerce and custom post type support, plus ACF field inclusion * Free: All llms.txt power controls — custom intro text, cache duration, pinning/ priority rules, taxonomy filtering, per-type post limits, and override mode * Free: llms-full.txt endpoint (full-content export per the llmstxt.org spec) * Free: Extended 90-day bot activity history, Insights tab, and bot management tab * Free: Wikipedia brand-presence advisory and the full AI Readiness dashboard * Fix: AI Readiness page could fail to load all pages due to an internal error — now lists every page correctly * Note: Pro now focuses on the features that rely on external AI/data services — AI Search Simulator, Competitor Intelligence, AI Coverage Gap (Google Search Console), and the firewall bot-access check #### 1.1.8 – 2026-05-05 * New: AI Coverage Gap dashboard (Pro) — connect Google Search Console to see pages with high Google impressions but zero AI bot visits, with HIGH/MED/LOW gap scoring * New: API Keys tab in Configuration — centralizes AI Simulator API keys and GSC OAuth connection in one place * New: Google Search Console OAuth integration via proxy relay — client secret stays server-side, one-click connect flow #### 1.1.7 – 2026-05-05 * Fix: Rank Math, Yoast, and AIOSEO compatibility — no longer outputs duplicate OG tags or Article JSON-LD when a major SEO plugin is active * Fix: llms-full.txt rules removed from robots.txt copy-paste block and auto-inject for free users (that URL returns a Pro-only error page) * Fix: Plugin no longer intercepts /llms.txt when Rank Math, Yoast, or AIOSEO is active — SEO plugin owns that URL * Fix: /.well-known/llms.txt now proxies the active SEO plugin’s /llms.txt content instead of serving a duplicate * Fix: Plugin activation no longer hard-flushes .htaccess rewrite rules, preventing conflicts with other plugins’ rewrite rules * New: AI discovery meta tags (llms-txt, ai-sitemap) now output on every page, not just singular posts * New: robots.txt health checks panel in AI Readiness dashboard — 5 checks including well-known/llms.txt accessibility, AI crawler rules, and Sitemap redirect detection * Fix: Inline JSON-LD validator in meta box now uses wp_add_inline_script() instead of a raw script tag * Fix: AI Readiness filter parameters (search, status, type, sort) now require a valid nonce — prevents unauthorized filtering * Fix: Competitor Intelligence page GET parameters now require a valid nonce * Fix: Referral tracker AJAX handler now verifies a nonce before processing POST data * Fix: JSON-LD schema output uses wp_json_encode() without flags that could bypass WordPress escaping rules * Fix: Database visit log query now passes LIMIT and OFFSET through wpdb::prepare() placeholders #### 1.1.6 – 2026-03-19 * Fix: Fatal error on Pages/Posts list — sfab_render_markdown_column used bare constant instead of string #### 1.1.5 – 2026-03-10 * Fix: AI Checklist meta description check now correctly detects Rank Math descriptions( including auto-generated from excerpt) * Fix: Bottom panel meta box now shows a “Refresh” notice after saving in Gutenberg so scores stay current * Fix: SEO Score check now shows your actual score (e.g. “14/100 via Rank Math”) alongside the fix recommendation #### 1.1.4 – 2026-03-09 * Fix: Fatal error in wp-admin — analytics-pages.php now loads before ai-readiness. php so ai_markdown_server_get_analytics_cache() is always defined #### 1.1.3 – 2026-03-05 * Fix: Fatal error on frontend pages — moved remote config functions to core-functions. php so they’re always available, not just on admin pages * Added error notification system: plugin emails the site admin on critical errors with rate-limiting (max once per hour per error type); falls back gracefully instead of crashing #### 1.1.2 – 2026-03-04 * Internal build fix: resolved premium package deployment issue #### 1.1.1 – 2026-03-04 * Improved AI Readiness page: rewrote help accordion sections in plain English with friendlier tone * Redesigned help cards to match site visual style (warm palette, serif headers, no colored borders) * Added ? trigger buttons on table headers to expand relevant help sections * Improved Sitemap tab: added help note explaining how to exclude pages from AI Markdown Sitemap; added smart link to active SEO plugin’s sitemap settings * Updated per-post checkbox label to “Include in AI discovery (llms.txt & ai-sitemap)” * Fixed h3 font size across admin pages (1.4rem) #### 1.1.0 – 2026-03-02 * Improved empty state UI: replaced emoji with custom robot logo on Bot Activity pages * Added dismissable helpful notice on AI Readiness page explaining that low-scoring pages may be intentional (login, password reset, dashboard pages) with suggestion to use noindex #### 1.0.3 – 2026-02-05 * Added Rule Analysis tools for Custom Handlers (Pro): Find Conflicts, Find Overlaps, Consolidate Rules * Improved rule management with automatic detection of duplicate and conflicting patterns #### 1.0.2 – 2026-02-04 * Improved Freemius uninstall integration for better user feedback tracking #### 1.0.1 – 2026-02-04 * Improved internationalization with proper translator comments * Code quality improvements for WordPress.org compliance * Updated to WordPress 6.9 compatibility #### 1.0.0 – 2026-02-04 Initial public release. **Content Serving** * Serve clean Markdown to AI crawlers via HTTP Content Negotiation*. md URL endpoints for any post (e.g., /your-post.md) * YAML frontmatter with metadata* Custom post type support * WooCommerce product integration **AI Discovery (llms.txt)** * Dual endpoints at /.well-known/llms.txt and /llms. txt * llms-full.txt comprehensive variant * 3-tier prioritization (SEO score > date > manual pins) * Smart filtering (word count, noindex, private content) * Pin rules for tags, categories, and specific posts **Bot Management** * 18+ curated AI bots (GPTBot, ClaudeBot, Perplexity, etc.) * Enable/disable individual bots * Bot visit analytics dashboard * Unknown bot tracking* Bot blocking via .htaccess **SEO Plugin Integration** * Auto-detect Yoast SEO, Rank Math, All In One SEO * Read SEO scores for prioritization * Respect noindex settings **Content Analysis** * AI Readiness scoring * Image alt text checker * Content structure analysis * Citation potential calculator * Competitor llms.txt analysis **Universal Content Handler** * Extract content from 20+ plugins (WPForms, Gravity Forms, Elementor, etc.) * Custom extraction rules with CSS/XPath selectors * FAQ pattern detection **Additional Features** * Onboarding wizard with site scan * AI Search Simulator to check visibility * AI referral tracking * Content freshness alerts * Per-post AI context customization * Extensibility filters for developers * Optional anonymous telemetry (disabled by default) ## Meta * Version **1.3.5** * Last updated **5 days ago** * Active installations **Fewer than 10** * WordPress version ** 5.6 or higher ** * Tested up to **7.1.3** * PHP version ** 7.4 or higher ** * Language * [English (US)](https://wordpress.org/plugins/seo-for-ai-bots/) * Tags * [AI Crawlers](https://pcd.wordpress.org/plugins/tags/ai-crawlers/)[ai seo](https://pcd.wordpress.org/plugins/tags/ai-seo/) [bot tracking](https://pcd.wordpress.org/plugins/tags/bot-tracking/)[ChatGPT](https://pcd.wordpress.org/plugins/tags/chatgpt/) [llms.txt](https://pcd.wordpress.org/plugins/tags/llms-txt/) * [Advanced View](https://pcd.wordpress.org/plugins/seo-for-ai-bots/advanced/) ## Ratings No reviews have been submitted yet. [Your review](https://wordpress.org/support/plugin/seo-for-ai-bots/reviews/#new-post) [See all reviews](https://wordpress.org/support/plugin/seo-for-ai-bots/reviews/) ## Contributors * [ thenooch ](https://profiles.wordpress.org/thenooch/) ## Support Got something to say? Need help? [View support forum](https://wordpress.org/support/plugin/seo-for-ai-bots/) ## Donate Would you like to support the advancement of this plugin? [ Donate to this plugin ](https://venmo.com/u/Daniel-Neumann-7)