Gemini 2.0 Flash, the former default, was shut down by Google; sites still using it resume on Gemini 3.6 Flash. Adds GPT-6, Claude Sonnet 5.5 and Opus 5.5 and Gemini 3.x, and warns before a selected model is retired.<\/p>","1.20.5":"
If your site uses Claude Sonnet 4 or Claude Opus 4.1, which Anthropic has retired, chats resume on their successors after this update. Also corrects Claude usage costs, which were overcounted for Haiku 4.5.<\/p>","1.20.4":"
The Site Learning screen no longer says it learns custom fields, which it has not done since 1.19.3.<\/p>","1.20.3":"
Stops an unrelated page being matched (and shown as a reference card) because of an everyday English word such as "need" in the question.<\/p>","1.20.2":"
Stops encrypted text appearing as a chat reply when Pro's response cache and message encryption are both on, and warns when the reCAPTCHA secret key can no longer be decrypted.<\/p>","1.20.1":"
Gives a clear message to a visitor whose browser is mistaken for a crawler, instead of the generic "currently unavailable" text.<\/p>","1.20.0":"
Adds an optional wp-config.php key so your API key and Pro license survive WordPress salt changes, and explains the cause when decryption does fail.<\/p>","1.19.7":"
Stops search-engine crawlers from creating fake conversations (and AI requests) by following deep-link question URLs.<\/p>","1.19.6":"
Fixes the background page being locked at the top on desktop when the chat is open, on sites with the optional "iOS keyboard fix" enabled.<\/p>","1.19.5":"
Fixes reference cards showing an unrelated page when the answer was grounded by vector search. Only affects sites with Vector Search enabled.<\/p>","1.19.4":"
Fixes System Health showing "Last crawl: Never run yet" while Site Learning was actually crawling. Each crawl run now records its own timestamp.<\/p>","1.19.3":"
Stops Site Learning from indexing third-party plugin metadata (e.g. Rank Math SEO fields) that could surface in chatbot answers. Re-run Site Learning after updating to clean already-indexed pages.<\/p>","1.19.2":"
Fixes the close (\u00d7) button not working when the chat is embedded on another site via the script. Recommended if you use the cross-site embed.<\/p>","1.19.1":"
Minor fix for the cross-site embed: the Google reCAPTCHA badge no longer covers the send button. Only relevant if you use reCAPTCHA and embed the chat on another site via the script.<\/p>","1.19.0":"
Large uploaded documents (TXT, MD, PDF, DOCX) are now split into parts automatically so each embeds on its own \u2014 no more cutting big files up by hand, and better answers from long documents. Existing entries are unchanged.<\/p>","1.18.1":"
Recommended bug-fix release. Fixes embedding that could keep sending requests to your AI provider without end \u2014 and, on a metered API key, keep incurring charges \u2014 when a document could not be embedded (too large, or a PDF with no extractable text). Also stops an oversized document from blocking the rest of a batch, shows why an entry failed to embed, and keeps the message avatar at its intended size. If you have uploaded large documents, update.<\/p>","1.18.0":"
Fixes a bug where embedding could keep sending requests to your AI provider without end \u2014 and, on a metered API key, keep incurring charges \u2014 when a document could not be embedded (too large, or a PDF with no extractable text). If you have uploaded large documents, update.<\/p>","1.17.0":"
The "Limit reached" usage message can now include a safe subset of HTML (links and basic emphasis) when you enable it, so you can send blocked visitors to a login or upgrade page. Off by default; existing setups are unchanged.<\/p>","1.16.2":"
Adds an optional text label beside the launcher icon (for example "Contact us here") so visitors know what the button does. Leave it empty to keep the icon-only button. New option only; existing setups are unchanged.<\/p>","1.16.1":"
Fixes the OpenAI-compatible provider (Qwen\/DashScope, DeepSeek, Zhipu): chat no longer reports "AI API key is not configured" when the key is set. Recommended for anyone using that provider.<\/p>","1.16.0":"
Adds an OpenAI-compatible provider (use Alibaba Qwen\/DashScope, DeepSeek, or Zhipu for both chat and RAG) and deep-link auto-send (a ?raplsaich_q= URL opens the chat and sends the question). New features only; existing setups are unchanged.<\/p>","1.15.4":"
Stops WordPress repeatedly offering an update you have already installed. Also a shorter plugin name. No behaviour changes.<\/p>","1.15.3":"
Fixes a stray "Choose file" upload control that could appear in the public chat widget on some themes. Recommended for all users.<\/p>","1.15.2":"
Fixes a JavaScript error that could appear as a chat message after the bot replied on sites running the Pro add-on. Recommended for all Pro users.<\/p>","1.15.0":"
Failed AI calls are now classified and visible to admins \u2014 an error badge in the conversation view and a 24-hour provider-error summary in System Health. Recommended for all users.<\/p>","1.14.1":"
Chat errors now show their real cause (API key, model, provider, or quota) instead of one generic message, and the custom quota message setting works again. Recommended for all users.<\/p>","1.14.0":"
Adds an "Add to knowledge" button on unanswered questions, a "Create draft page" button for generated FAQs, and a dashboard System Health panel. Recommended for all users.<\/p>","1.13.0":"
Adds visitor-initiated chat history deletion (privacy), an Unanswered Questions report, a one-click FAQ draft generator, optional GA4 events, and WP-CLI commands. All new data features are off by default. Recommended for all users.<\/p>","1.12.0":"
Adds a setup checklist, a demo preview, industry starter templates, page context, a monthly Cost Guard, model fallback, and an optional weekly summary email. Recommended for all users.<\/p>","1.11.1":"
Blocked messages now say which limit was reached (quota, credits, token, daily, or rate limit). Recommended for sites using Usage Control.<\/p>","1.11.0":"
Adds optional "Grounded Answers Only" (anti-hallucination) and "Usage Control" (per-visitor\/user caps to protect your API spend). Both off by default. Recommended for all users.<\/p>","1.10.0":"
Adds an optional "AI-smell score" for Japanese bot replies in the Conversations log (detection only; replies are never changed). Recommended for all users.<\/p>","1.9.4":"
Feedback buttons (\ud83d\udc4d\ud83d\udc4e) on bot messages now default to off for new installs. Existing settings are unchanged.<\/p>","1.9.3":"
Adds support for Google's new "AQ." Gemini API keys (the legacy "AIza" format is being retired by Google) and warns if your current Gemini key needs migrating. Recommended for all Gemini users.<\/p>","1.9.2":"
The free onboarding now lets you choose between OpenRouter free models and the Google Gemini free tier, each with an up-front data-handling note. Recommended for all users.<\/p>","1.9.1":"
Fixes a chat freeze when AI responses contain markdown tables, and API key deletion on WordPress 7.0 sites. Recommended for all users.<\/p>","1.9.0":"
Adds a one-minute onboarding flow that lets new users connect a free OpenRouter API key (no credit card) and start chatting immediately. Includes automatic fallback when free models are rate-limited upstream.<\/p>","1.8.2":"
Documentation refresh \u2014 updated description, features, and FAQ. No functional changes.<\/p>","1.8.1":"
Fixes a PHP "Undefined array key" warning in the chatbot widget. Recommended for all.<\/p>","1.8.0":"
Adds the WordPress 7.0 "AI Client (Connectors)" provider option. Recommended for WordPress 7.0 sites.<\/p>","1.5.0":"
Major release: Gutenberg block, Abilities API, language auto-detect. Recommended for all users.<\/p>"},"ratings":{"1":0,"2":0,"3":0,"4":1,"5":8},"assets_icons":{"icon-128x128.png":{"filename":"icon-128x128.png","revision":3493381,"resolution":"128x128","location":"assets","locale":"","width":128,"height":128},"icon-256x256.png":{"filename":"icon-256x256.png","revision":3493381,"resolution":"256x256","location":"assets","locale":"","width":256,"height":256}},"assets_banners":{"banner-1544x500.png":{"filename":"banner-1544x500.png","revision":3493381,"resolution":"1544x500","location":"assets","locale":"","width":1544,"height":500},"banner-772x250.png":{"filename":"banner-772x250.png","revision":3493381,"resolution":"772x250","location":"assets","locale":"","width":772,"height":250}},"assets_blueprints":{},"all_blocks":{"rapls-ai-chatbot\/chatbot":{"$schema":"https:\/\/schemas.wp.org\/trunk\/block.json","apiVersion":3,"name":"rapls-ai-chatbot\/chatbot","version":"1.5.0","title":"AI Chatbot","category":"widgets","icon":"format-chat","description":"Embed an AI chatbot inline on your page.","keywords":["chat","ai","chatbot","assistant","rapls"],"supports":{"html":false,"multiple":false},"attributes":{"height":{"type":"string","default":"500px"},"theme":{"type":"string","default":""},"bot":{"type":"string","default":""}},"textdomain":"rapls-ai-chatbot","editorScript":"file:.\/editor.js","render":"file:.\/render.php"}},"tagged_versions":["1.10.0","1.11.1","1.14.0","1.14.1","1.15.0","1.15.1","1.15.2","1.15.3","1.15.4","1.16.0","1.16.1","1.16.2","1.17.0","1.18.1","1.19.2","1.19.3","1.19.4","1.19.6","1.19.7","1.20.0","1.20.1","1.20.2","1.20.3","1.20.4","1.20.5","1.21.0","1.5.10","1.5.11","1.5.12","1.5.13","1.5.14","1.5.18","1.5.20","1.5.21","1.5.6","1.5.7","1.5.9","1.6.4","1.6.5","1.7.1","1.7.5","1.7.6","1.7.7","1.8.0","1.8.1","1.8.2","1.9.0","1.9.1","1.9.2","1.9.3","1.9.4"],"block_files":[],"assets_screenshots":{"screenshot-1.png":{"filename":"screenshot-1.png","revision":3594068,"resolution":"1","location":"assets","locale":"","width":800,"height":1074},"screenshot-10.png":{"filename":"screenshot-10.png","revision":3597591,"resolution":"10","location":"assets","locale":"","width":2412,"height":1204},"screenshot-2.png":{"filename":"screenshot-2.png","revision":3597591,"resolution":"2","location":"assets","locale":"","width":1332,"height":1682},"screenshot-3.png":{"filename":"screenshot-3.png","revision":3597591,"resolution":"3","location":"assets","locale":"","width":1832,"height":1684},"screenshot-4.png":{"filename":"screenshot-4.png","revision":3493381,"resolution":"4","location":"assets","locale":"","width":1280,"height":800},"screenshot-5.png":{"filename":"screenshot-5.png","revision":3493381,"resolution":"5","location":"assets","locale":"","width":1286,"height":800},"screenshot-6.png":{"filename":"screenshot-6.png","revision":3594068,"resolution":"6","location":"assets","locale":"","width":1280,"height":850},"screenshot-7.png":{"filename":"screenshot-7.png","revision":3594068,"resolution":"7","location":"assets","locale":"","width":1280,"height":860},"screenshot-8.png":{"filename":"screenshot-8.png","revision":3594068,"resolution":"8","location":"assets","locale":"","width":1269,"height":800},"screenshot-9.png":{"filename":"screenshot-9.png","revision":3597591,"resolution":"9","location":"assets","locale":"","width":2432,"height":1296}},"screenshots":{"1":"Chat widget \u2014 the front-end chatbot answering a visitor from your site content.","2":"Start for free \u2014 the onboarding panel connects a no-credit-card OpenRouter or Google Gemini free key in about a minute, with each option's data-handling trade-off shown up front.","3":"Dashboard \u2014 setup checklist, conversation\/index\/knowledge stats, System Health panel (API key, cron, REST reachability), and 30-day token usage with an estimated cost.","4":"Knowledge base \u2014 add custom text\/Q&A or import TXT, CSV, MD, PDF, or DOCX files.","5":"Conversations \u2014 review, filter, archive, and export visitor conversations with lead and status details.","6":"Analytics \u2014 conversation, message, cost, and quality metrics (Pro).","7":"Site Learning \u2014 auto-crawl posts, pages, and custom post types so the bot answers from your own content.","8":"AI Settings \u2014 choose your provider (OpenAI, Anthropic Claude, Google Gemini, or OpenRouter), enter your API key, pick a model, and enable Vector Embedding (RAG).","9":"Unanswered Questions \u2014 questions the bot could not answer from your content, with one-click \"Add to knowledge\" and an AI-generated FAQ draft you can save as a draft page.","10":"Cost controls \u2014 Monthly Cost Guard with budget warnings, per-visitor usage limits, and model fallback so API bills never run away."}},"plugin_section":[],"plugin_tags":[191735,2364,216196,242115,70446],"plugin_category":[],"plugin_contributors":[253146],"plugin_business_model":[],"class_list":["post-292250","plugin","type-plugin","status-publish","hentry","plugin_tags-ai-chatbot","plugin_tags-chatbot","plugin_tags-chatgpt","plugin_tags-mcp","plugin_tags-rag","plugin_contributors-rapls","plugin_committers-rapls"],"banners":{"banner":"https:\/\/ps.w.org\/rapls-ai-chatbot\/assets\/banner-772x250.png?rev=3493381","banner_2x":"https:\/\/ps.w.org\/rapls-ai-chatbot\/assets\/banner-1544x500.png?rev=3493381","banner_rtl":false,"banner_2x_rtl":false},"icons":{"svg":false,"icon":"https:\/\/ps.w.org\/rapls-ai-chatbot\/assets\/icon-128x128.png?rev=3493381","icon_2x":"https:\/\/ps.w.org\/rapls-ai-chatbot\/assets\/icon-256x256.png?rev=3493381","generated":false},"screenshots":[{"src":"https:\/\/ps.w.org\/rapls-ai-chatbot\/assets\/screenshot-1.png?rev=3594068","caption":"Chat widget \u2014 the front-end chatbot answering a visitor from your site content."},{"src":"https:\/\/ps.w.org\/rapls-ai-chatbot\/assets\/screenshot-2.png?rev=3597591","caption":"Start for free \u2014 the onboarding panel connects a no-credit-card OpenRouter or Google Gemini free key in about a minute, with each option's data-handling trade-off shown up front."},{"src":"https:\/\/ps.w.org\/rapls-ai-chatbot\/assets\/screenshot-3.png?rev=3597591","caption":"Dashboard \u2014 setup checklist, conversation\/index\/knowledge stats, System Health panel (API key, cron, REST reachability), and 30-day token usage with an estimated cost."},{"src":"https:\/\/ps.w.org\/rapls-ai-chatbot\/assets\/screenshot-4.png?rev=3493381","caption":"Knowledge base \u2014 add custom text\/Q&A or import TXT, CSV, MD, PDF, or DOCX files."},{"src":"https:\/\/ps.w.org\/rapls-ai-chatbot\/assets\/screenshot-5.png?rev=3493381","caption":"Conversations \u2014 review, filter, archive, and export visitor conversations with lead and status details."},{"src":"https:\/\/ps.w.org\/rapls-ai-chatbot\/assets\/screenshot-6.png?rev=3594068","caption":"Analytics \u2014 conversation, message, cost, and quality metrics (Pro)."},{"src":"https:\/\/ps.w.org\/rapls-ai-chatbot\/assets\/screenshot-7.png?rev=3594068","caption":"Site Learning \u2014 auto-crawl posts, pages, and custom post types so the bot answers from your own content."},{"src":"https:\/\/ps.w.org\/rapls-ai-chatbot\/assets\/screenshot-8.png?rev=3594068","caption":"AI Settings \u2014 choose your provider (OpenAI, Anthropic Claude, Google Gemini, or OpenRouter), enter your API key, pick a model, and enable Vector Embedding (RAG)."},{"src":"https:\/\/ps.w.org\/rapls-ai-chatbot\/assets\/screenshot-9.png?rev=3597591","caption":"Unanswered Questions \u2014 questions the bot could not answer from your content, with one-click \"Add to knowledge\" and an AI-generated FAQ draft you can save as a draft page."},{"src":"https:\/\/ps.w.org\/rapls-ai-chatbot\/assets\/screenshot-10.png?rev=3597591","caption":"Cost controls \u2014 Monthly Cost Guard with budget warnings, per-visitor usage limits, and model fallback so API bills never run away."}],"raw_content":"\n
Add an AI chatbot that answers visitors from your own<\/em> content \u2014 and admits when it doesn't know.<\/strong> Rapls AI Chatbot searches your posts, pages, and knowledge base first, then replies in natural language, in the visitor's own language. With Grounded Answers Only mode, if your content has no answer, the bot says so instead of making something up.<\/p>\n\n You can be live in about five minutes, for free, with no credit card.<\/strong> On first setup a guided \"Start for free\" panel hands you a free key from OpenRouter or the Google Gemini free tier, tests it, auto-selects a working model, and switches the chatbot on \u2014 no AI or API experience needed.<\/p>\n\n Watch the setup, start to finish:<\/strong><\/p>\n\n https:\/\/www.youtube.com\/watch?v=6mFve1lxvuw<\/p>\n\n No monthly SaaS fee \u2014 and no runaway bill.<\/strong> It runs on your own API key with no markup, and built-in Usage Control caps daily usage per visitor, so you can open the bot to the public without worrying about your API spend. A Monthly Cost Guard lets you set a hard budget, and Model Fallback keeps the bot answering on a lightweight model when your main model hits its quota. Conversations and keys stay on your own server.<\/p>\n\n Built for site owners, agencies, and developers who want control over the model, the data, and the cost. Connect OpenAI, Anthropic Claude, Google Gemini, or OpenRouter, or any OpenAI-compatible endpoint such as Alibaba Tongyi Qwen (DashScope) \u2014 useful where OpenAI and Gemini are hard to reach \u2014 and switch anytime.<\/p>\n\n The bot doesn't just answer \u2014 it tells you what your site is missing. Questions it could not answer from your content are recorded on the dashboard, one click turns them into a knowledge entry, and another click drafts an FAQ page from the last 30 days of real visitor questions. Answer gaps become content, content improves the bot, and the same pages work for SEO and AI search (LLMO).<\/p>\n\n Conversations and keys stay on your own WordPress install. You are billed by your AI provider directly, so cost is transparent and there is no markup. Visitors can delete their own chat history from the widget (optional), and retention is configurable \u2014 GDPR-friendly by design.<\/p>\n\n Free \u2014 not a trial.<\/strong> The free version is the full product: chatbot, site learning, knowledge base, RAG hybrid search, Grounded Answers Only, MCP server (5 tools), a usage dashboard (conversations, messages, and cost), and per-visitor daily caps. Run it forever at no cost on a free-tier key.<\/p>\n\n Pro \u2014 for sites that run the bot as a business tool:<\/strong><\/p>\n\n You do not need an API key or any AI experience to start. On first setup the plugin shows a \"Start for free\" panel with two no-credit-card paths: an OpenRouter free key or a Google Gemini free-tier key.<\/p>\n\n That's it. Each option states its data-handling trade-off up front (the Gemini free tier may use submitted content to improve Google's models), so you choose with eyes open. You can switch to your own OpenAI, Claude, or Gemini key at any time.<\/p>\n\n Learn more: Plugin details<\/a> | Source code (GitHub)<\/a><\/p>\n\n Watch the full setup \u2014 from install to first reply \u2014 in about 5 minutes, no credit card required:<\/p>\n\n https:\/\/www.youtube.com\/watch?v=KWEeYuZ0uEg<\/p>\n\n See it answer from your own content, and how that differs from plain ChatGPT:<\/p>\n\n https:\/\/www.youtube.com\/watch?v=HgbYr6c_QlI<\/p>\n\n This plugin connects to the following external third-party services. No data is sent to any service until you configure an API key and enable the feature in the plugin settings.<\/strong> Each service requires the site administrator to create an account and obtain API credentials. By using these services, you agree to their respective terms and privacy policies listed below.<\/p>\n\n Used when you select OpenAI as your AI provider. User messages and optionally site content are sent to generate AI responses.<\/p>\n\n Used when you select Anthropic Claude as your AI provider. User messages and optionally site content are sent to generate AI responses.<\/p>\n\n Used when you select Google Gemini as your AI provider. User messages and optionally site content are sent to generate AI responses.<\/p>\n\n Used when you select OpenRouter as your AI provider. OpenRouter is a unified API gateway that routes requests to various AI models.<\/p>\n\n Used only if you select the \"OpenAI-compatible\" provider and enter a base URL. User messages, and optionally your site content (including indexed content for RAG embeddings), are sent to the endpoint you configure to generate responses or embeddings. Because the endpoint is chosen by you, please review the terms and privacy policy of the specific provider you connect to. For example, Alibaba Cloud Model Studio (DashScope), which powers Tongyi Qwen:<\/p>\n\n Used only if you enable reCAPTCHA in the plugin settings for spam protection. The visitor's IP address and interaction data are sent to Google for verification.<\/p>\n\n Used only if you enable the LINE integration via the Pro add-on. Connects to the LINE Messaging API for chatbot-to-LINE messaging.<\/p>\n\n The plugin includes an optional embed loader script ( You can disable these features in the plugin settings:\n* Conversation history saving\n* Site content crawling\/learning\n* Google reCAPTCHA verification\n* Web search<\/p>\n\n\n \ud83d\udc49 Plugin details:<\/strong> Rapls AI Chatbot<\/a><\/p>\n\n\n The API key (and the Pro license) is encrypted with a key derived from your WordPress security salts (AUTH_KEY \/ AUTH_SALT). If something changes those salts, the stored key can no longer be read. WordPress uses salt values kept in the database whenever the wp-config.php constants are missing, still set to \"put your unique phrase here\", or reused across constants \u2014 and those database values can be regenerated without wp-config.php changing. Security plugins that rotate salts, a failing object cache, and restored database backups are the usual causes; you will normally be logged out of WordPress at the same time.<\/p>\n\n To make the stored keys independent of the salts, add a long random string to wp-config.php, above the \"That's all, stop editing!\" line:<\/p>\n\n Existing keys are re-encrypted with it automatically the next time you open wp-admin, and they then survive any salt change. Keep the value backed up: losing it means re-entering your API key and license once.<\/p><\/dd>\n Yes \u2014 the plugin is free, and the onboarding panel connects a no-credit-card OpenRouter or Google Gemini free-tier key in about a minute. Free tiers have rate limits, so for production traffic you may want your own provider key.<\/p><\/dd>\n Yes. Rapls runs on your own API key so your data and cost stay under your control. If you do not have one, the plugin guides you to a free OpenRouter key or a free Google Gemini key on first setup \u2014 both need no credit card.<\/p><\/dd>\n Not if you don't want it to. Turn on Grounded Answers Only and the bot replies \"I couldn't find that on this site\" whenever your content has no relevant answer, instead of answering from the model's general knowledge.<\/p><\/dd>\n No. Usage Control caps daily usage per guest and per logged-in user on the server side, separate from rate limiting. Pro adds role-based limits and per-user credits.<\/p><\/dd>\n Yes. The free version ships an MCP (Model Context Protocol) server with 5 tools, so AI agents such as Claude or ChatGPT can search your posts, read knowledge entries, and act on your site through conversation. Pro adds product-search and analytics tools on top.<\/p><\/dd>\n It depends on the provider you choose, and the plugin tells you before you pick. OpenRouter free models are served by various upstream providers, each with its own data-handling policy. The Google Gemini free tier may use your submitted content to improve Google's models \u2014 if you do not want that, use a paid Gemini tier or another provider. Either way, your conversations and keys are stored on your own WordPress install, not on Rapls servers.<\/p><\/dd>\n Yes. Google is moving Gemini API keys from the legacy \"AIza\" standard format to the new \"AQ.\" auth format, and Google AI Studio now issues \"AQ.\" keys by default. Rapls works with both. Google is retiring the standard format (unrestricted \"AIza\" keys are rejected from 2026-06-19, and all \"AIza\" keys from 2026-09), so if you still have an \"AIza\" key the plugin shows a notice with how to migrate. New keys created today are already \"AQ.\" keys and need no action.<\/p><\/dd>\n Yes. Configure multiple providers in Settings and switch between them. WordPress 7.0 Connectors API also supports unified key management.<\/p><\/dd>\n Yes. Enable \"Site Learning\" in Settings to crawl published content (posts, pages, custom post types, WooCommerce products). Configure crawl scope and frequency.<\/p><\/dd>\n Yes. Configure cross-site embed in Display Settings. Use iframe or script loader.<\/p><\/dd>\n In Settings > AI Settings, Connectors UI appears if WP 7.0 is active. Register your Claude API key once; all Connectors-compatible plugins access it.<\/p><\/dd>\n Yes. Data Management tab lets you save\/review all conversations. Configure retention period (30\/90\/365 days or indefinite).<\/p><\/dd>\n See Plugin details<\/a> or WordPress.org Support<\/a><\/p><\/dd>\n\n<\/dl>\n\n\nWhy site owners pick Rapls<\/h4>\n\n
\n
What it does<\/h4>\n\n
\n
?raplsaich_q=<\/code> URL opens the chatbot and sends the question automatically, so end-of-article \"suggested questions\" become one-click conversations.<\/li>\n<\/ul>\n\nTurns visitor questions into content<\/h4>\n\n
Self-hosted and private<\/h4>\n\n
Free and Pro<\/h4>\n\n
\n
Up and running in about 5 minutes (free, no credit card)<\/h4>\n\n
\n
See it in action<\/h4>\n\n
External Services<\/h3>\n\n
1. OpenAI (GPT models): AI Provider<\/h4>\n\n
\n
2. Anthropic (Claude models): AI Provider<\/h4>\n\n
\n
3. Google (Gemini models): AI Provider<\/h4>\n\n
\n
4. OpenRouter: AI Provider<\/h4>\n\n
\n
5. OpenAI-compatible endpoints (e.g. Alibaba Qwen\/DashScope, DeepSeek, Zhipu): AI Provider<\/h4>\n\n
\n
6. Google reCAPTCHA v3 (Optional)<\/h4>\n\n
\n
7. LINE Messaging API (Pro Add-on, Optional)<\/h4>\n\n
\n
Cross-Site Embed<\/h4>\n\n
embed-loader.js<\/code>) for embedding the chatbot on external websites via an iframe. This script does not load any external CDN resources or third-party scripts. It creates an iframe pointing back to your own WordPress site, and all data processing occurs on your server.<\/p>\n\nData Transmitted to External Services<\/h4>\n\n
\n
Data Storage<\/h4>\n\n
\n
User Controls<\/h4>\n\n
\n
rapls-ai-chatbot<\/code> folder to \/wp-content\/plugins\/<\/code><\/li>\nGetting Started<\/h4>\n\n
\n
\n
My API key stops working every few days and I have to re-enter it<\/h3><\/dt>\n
define('RAPLSAICH_ENCRYPTION_KEY', 'a-long-random-string-of-at-least-32-characters');\n<\/code><\/pre>\n\nCan I use it for free?<\/h3><\/dt>\n
Do I need an API key?<\/h3><\/dt>\n
Will the bot make things up?<\/h3><\/dt>\n
Can visitors run up my API bill?<\/h3><\/dt>\n
Does it work with AI agents (MCP)?<\/h3><\/dt>\n
Is my data private on the free options?<\/h3><\/dt>\n
My Gemini key starts with \"AQ.\" instead of \"AIza\" \u2014 is that OK?<\/h3><\/dt>\n
Can I use multiple AI providers?<\/h3><\/dt>\n
Does it crawl my entire site automatically?<\/h3><\/dt>\n
Can I embed on external sites?<\/h3><\/dt>\n
How do I set up WordPress 7.0 Connectors API?<\/h3><\/dt>\n
Is there conversation history?<\/h3><\/dt>\n
Troubleshooting: chat not appearing?<\/h3><\/dt>\n
\n
More questions?<\/h3><\/dt>\n
1.21.0<\/h4>\n\n
\n
1.20.5<\/h4>\n\n
\n
1.20.4<\/h4>\n\n
\n
1.20.3<\/h4>\n\n
\n
1.20.2<\/h4>\n\n
\n
1.20.1<\/h4>\n\n
\n
bot_request<\/code> error code had no entry in the message map. It now says the request looked automated and who to contact, which matters for the rare visitor whose browser is misidentified.<\/li>\n<\/ul>\n\n1.20.0<\/h4>\n\n
\n
RAPLSAICH_ENCRYPTION_KEY<\/code> \u2014 an optional wp-config.php constant that decouples the stored API key and Pro license from the WordPress security salts. Sites whose salts are rotated (by a security plugin, a failing object cache, or a restored database) lost both every few days and had to re-enter them. Define the constant and existing keys are re-encrypted with it on the next wp-admin request; they then survive any salt change. Nothing changes for sites that do not define it. See the FAQ for details.<\/li>\n1.19.7<\/h4>\n\n
\n
?raplsaich_q=\u2026<\/code>, the question was sent automatically, so every crawl showed up under Conversations as a new visitor and triggered a real AI request. Each crawl gets a new session, so per-visitor limits did not stop it. Known crawlers and automated browsers no longer auto-send deep-link questions, and the server refuses chat requests from crawler User-Agents before the AI is called or anything is saved. Developers can adjust detection with the new raplsaich_is_bot_request filter. Thanks to xyp for the detailed report.<\/li>\n<\/ul>\n\n1.19.6<\/h4>\n\n
\n
1.19.5<\/h4>\n\n
\n
1.19.4<\/h4>\n\n
\n
1.19.3<\/h4>\n\n
\n
1.19.2<\/h4>\n\n
\n
1.19.1<\/h4>\n\n
\n
1.19.0<\/h4>\n\n
\n
1.18.1<\/h4>\n\n
\n
1.18.0<\/h4>\n\n
\n
1.17.0<\/h4>\n\n
\n
, , ); everything else is stripped on save, and the widget scrubs the message again before rendering. Off by default, so existing setups are unchanged.<\/em><\/strong><\/a><\/li>\n<\/ul>\n\n1.16.2<\/h4>\n\n
\n
1.16.1<\/h4>\n\n
\n
1.16.0<\/h4>\n\n
\n
?raplsaich_q=your+question<\/code> (a plain ?q=<\/code> also works) opens the chatbot and sends that question automatically, with no typing or copy-paste \u2014 ideal for turning end-of-article \"suggested questions\" into one-click conversations. The question is shown as a normal visitor message, and the parameter is removed from the URL after sending so a refresh does not resend it.<\/li>\n<\/ul>\n\n