{"id":254489,"date":"2025-10-06T20:23:29","date_gmt":"2025-10-06T20:23:29","guid":{"rendered":"https:\/\/wordpress.org\/plugins\/ai-chat\/"},"modified":"2026-04-02T07:25:39","modified_gmt":"2026-04-02T07:25:39","slug":"axiachat-ai","status":"publish","type":"plugin","link":"https:\/\/id.wordpress.org\/plugins\/axiachat-ai\/","author":23219905,"comment_status":"closed","ping_status":"closed","template":"","meta":{"_crdt_document":"","version":"4.1.1","stable_tag":"4.1.1","tested":"6.9.4","requires":"5.0","requires_php":"7.4","requires_plugins":null,"header_name":"AxiaChat AI \u2013 Free AI Chatbot (Answers Customers Automatically)","header_author":"estebandezafra","header_description":"A customizable AI chatbot for WordPress with contextual embeddings, multi\u2011provider support and upcoming action rules.","assets_banners_color":"476280","last_updated":"2026-04-02 07:25:39","external_support_url":"","external_repository_url":"","donate_link":"","header_plugin_uri":"https:\/\/axiachat.org","header_author_uri":"https:\/\/axiachat.org","rating":5,"author_block_rating":0,"active_installs":1000,"downloads":16446,"num_ratings":12,"support_threads":6,"support_threads_resolved":6,"author_block_count":0,"sections":["installation","faq","changelog","description"],"tags":{"1.1.5":{"tag":"1.1.5","author":"estebandezafra","date":"2025-10-06 20:33:25"},"1.1.6":{"tag":"1.1.6","author":"estebandezafra","date":"2025-10-12 11:30:28"},"1.1.7":{"tag":"1.1.7","author":"estebandezafra","date":"2025-10-21 09:10:26"},"1.1.8":{"tag":"1.1.8","author":"estebandezafra","date":"2025-10-17 06:48:31"},"1.1.9":{"tag":"1.1.9","author":"estebandezafra","date":"2025-10-21 09:11:41"},"1.2.0":{"tag":"1.2.0","author":"estebandezafra","date":"2025-10-22 13:49:11"},"1.2.1":{"tag":"1.2.1","author":"estebandezafra","date":"2025-10-24 16:35:31"},"1.2.3":{"tag":"1.2.3","author":"estebandezafra","date":"2025-11-02 08:43:33"},"1.2.5":{"tag":"1.2.5","author":"estebandezafra","date":"2025-11-09 11:20:44"},"1.2.6":{"tag":"1.2.6","author":"estebandezafra","date":"2025-11-16 08:04:28"},"1.2.7":{"tag":"1.2.7","author":"estebandezafra","date":"2025-12-05 12:10:14"},"1.2.8":{"tag":"1.2.8","author":"estebandezafra","date":"2026-01-15 15:24:32"},"1.2.9":{"tag":"1.2.9","author":"estebandezafra","date":"2026-01-21 14:44:40"},"1.7.0":{"tag":"1.7.0","author":"estebandezafra","date":"2025-10-14 09:57:24"},"2.0.5":{"tag":"2.0.5","author":"estebandezafra","date":"2026-02-08 22:32:19"},"2.0.6":{"tag":"2.0.6","author":"estebandezafra","date":"2026-02-10 07:35:54"},"2.0.7":{"tag":"2.0.7","author":"estebandezafra","date":"2026-02-11 05:24:37"},"2.0.8":{"tag":"2.0.8","author":"estebandezafra","date":"2026-02-12 08:06:58"},"2.0.9":{"tag":"2.0.9","author":"estebandezafra","date":"2026-02-13 06:17:51"},"3.0.1":{"tag":"3.0.1","author":"estebandezafra","date":"2026-02-16 21:11:02"},"3.0.2":{"tag":"3.0.2","author":"estebandezafra","date":"2026-02-20 05:34:19"},"3.0.3":{"tag":"3.0.3","author":"estebandezafra","date":"2026-02-28 16:10:20"},"3.0.4":{"tag":"3.0.4","author":"estebandezafra","date":"2026-02-28 21:02:17"},"3.0.5":{"tag":"3.0.5","author":"estebandezafra","date":"2026-03-03 07:34:33"},"3.0.6":{"tag":"3.0.6","author":"estebandezafra","date":"2026-03-03 18:49:19"},"3.0.7":{"tag":"3.0.7","author":"estebandezafra","date":"2026-03-04 11:08:20"},"3.0.8":{"tag":"3.0.8","author":"estebandezafra","date":"2026-03-05 11:48:04"},"3.0.9":{"tag":"3.0.9","author":"estebandezafra","date":"2026-03-07 16:09:14"},"3.1.5":{"tag":"3.1.5","author":"estebandezafra","date":"2026-03-11 10:35:49"},"4.0.2":{"tag":"4.0.2","author":"estebandezafra","date":"2026-03-14 16:57:01"},"4.1.0":{"tag":"4.1.0","author":"estebandezafra","date":"2026-03-27 12:22:20"},"4.1.1":{"tag":"4.1.1","author":"estebandezafra","date":"2026-04-02 07:25:39"}},"upgrade_notice":[],"ratings":{"1":0,"2":0,"3":0,"4":0,"5":12},"assets_icons":{"icon-256x256.png":{"filename":"icon-256x256.png","revision":3412166,"resolution":"256x256","location":"assets","locale":""}},"assets_banners":{"banner-1544x500.png":{"filename":"banner-1544x500.png","revision":3459630,"resolution":"1544x500","location":"assets","locale":""},"banner-772x250.png":{"filename":"banner-772x250.png","revision":3459630,"resolution":"772x250","location":"assets","locale":""}},"assets_blueprints":{},"all_blocks":[],"tagged_versions":["1.1.5","1.1.6","1.1.7","1.1.8","1.1.9","1.2.0","1.2.1","1.2.3","1.2.5","1.2.6","1.2.7","1.2.8","1.2.9","1.7.0","2.0.5","2.0.6","2.0.7","2.0.8","2.0.9","3.0.1","3.0.2","3.0.3","3.0.4","3.0.5","3.0.6","3.0.7","3.0.8","3.0.9","3.1.5","4.0.2","4.1.0","4.1.1"],"block_files":[],"assets_screenshots":{"screenshot-1.png":{"filename":"screenshot-1.png","revision":3473922,"resolution":"1","location":"assets","locale":""},"screenshot-2.png":{"filename":"screenshot-2.png","revision":3473922,"resolution":"2","location":"assets","locale":""},"screenshot-3.png":{"filename":"screenshot-3.png","revision":3473922,"resolution":"3","location":"assets","locale":""},"screenshot-4.png":{"filename":"screenshot-4.png","revision":3473922,"resolution":"4","location":"assets","locale":""},"screenshot-5.png":{"filename":"screenshot-5.png","revision":3473922,"resolution":"5","location":"assets","locale":""},"screenshot-6.png":{"filename":"screenshot-6.png","revision":3473922,"resolution":"6","location":"assets","locale":""},"screenshot-7.png":{"filename":"screenshot-7.png","revision":3473922,"resolution":"7","location":"assets","locale":""}},"screenshots":{"1":"<strong>Bot Configuration<\/strong> \u2013 Set the AI model, behavior, tone, and appearance for each chatbot independently","2":"<strong>Training Context<\/strong> \u2013 Index your WordPress content, WooCommerce products, and PDFs for accurate RAG responses","3":"<strong>Training Instructions<\/strong> \u2013 Write or AI-generate the bot\u2019s personality and system prompt with one click","4":"<strong>Easy Setup Wizard<\/strong> \u2013 Launch a fully working AI chatbot in under 2 minutes with a guided step-by-step flow","5":"<strong>Lead Capture Dashboard<\/strong> \u2013 View, filter, and export leads collected automatically by the chatbot during conversations","6":"<strong>Usage &amp; Cost Tracking<\/strong> \u2013 Monitor token consumption, API cost per conversation, and daily usage in real time","7":"<strong>Appointment Management<\/strong> \u2013 Review and manage AI-scheduled appointments directly from the WordPress admin panel"},"jetpack_post_was_ever_published":false},"plugin_section":[],"plugin_tags":[191735,216196,2369,237277,286],"plugin_category":[41],"plugin_contributors":[238115],"plugin_business_model":[],"class_list":["post-254489","plugin","type-plugin","status-publish","hentry","plugin_tags-ai-chatbot","plugin_tags-chatgpt","plugin_tags-customer-support","plugin_tags-google-gemini","plugin_tags-woocommerce","plugin_category-communication","plugin_contributors-estebandezafra","plugin_committers-estebandezafra"],"banners":{"banner":"https:\/\/ps.w.org\/axiachat-ai\/assets\/banner-772x250.png?rev=3459630","banner_2x":"https:\/\/ps.w.org\/axiachat-ai\/assets\/banner-1544x500.png?rev=3459630","banner_rtl":false,"banner_2x_rtl":false},"icons":{"svg":false,"icon":"https:\/\/ps.w.org\/axiachat-ai\/assets\/icon-256x256.png?rev=3412166","icon_2x":"https:\/\/ps.w.org\/axiachat-ai\/assets\/icon-256x256.png?rev=3412166","generated":false},"screenshots":[{"src":"https:\/\/ps.w.org\/axiachat-ai\/assets\/screenshot-1.png?rev=3473922","caption":"<strong>Bot Configuration<\/strong> \u2013 Set the AI model, behavior, tone, and appearance for each chatbot independently"},{"src":"https:\/\/ps.w.org\/axiachat-ai\/assets\/screenshot-2.png?rev=3473922","caption":"<strong>Training Context<\/strong> \u2013 Index your WordPress content, WooCommerce products, and PDFs for accurate RAG responses"},{"src":"https:\/\/ps.w.org\/axiachat-ai\/assets\/screenshot-3.png?rev=3473922","caption":"<strong>Training Instructions<\/strong> \u2013 Write or AI-generate the bot\u2019s personality and system prompt with one click"},{"src":"https:\/\/ps.w.org\/axiachat-ai\/assets\/screenshot-4.png?rev=3473922","caption":"<strong>Easy Setup Wizard<\/strong> \u2013 Launch a fully working AI chatbot in under 2 minutes with a guided step-by-step flow"},{"src":"https:\/\/ps.w.org\/axiachat-ai\/assets\/screenshot-5.png?rev=3473922","caption":"<strong>Lead Capture Dashboard<\/strong> \u2013 View, filter, and export leads collected automatically by the chatbot during conversations"},{"src":"https:\/\/ps.w.org\/axiachat-ai\/assets\/screenshot-6.png?rev=3473922","caption":"<strong>Usage &amp; Cost Tracking<\/strong> \u2013 Monitor token consumption, API cost per conversation, and daily usage in real time"},{"src":"https:\/\/ps.w.org\/axiachat-ai\/assets\/screenshot-7.png?rev=3473922","caption":"<strong>Appointment Management<\/strong> \u2013 Review and manage AI-scheduled appointments directly from the WordPress admin panel"}],"raw_content":"<!--section=installation-->\n<ol>\n<li>Upload the plugin files to the <code>\/wp-content\/plugins\/axiachat-ai<\/code> directory, or install through the WordPress plugin screen.<\/li>\n<li>Activate the plugin through the 'Plugins' screen in WordPress.<\/li>\n<li>Run the Easy Setup Wizard.<\/li>\n<li>Add your AI provider API key.<\/li>\n<li>Test your chatbot using the built-in preview \u2014 click \"Preview Bot\" in the bot settings to see it live before deploying globally.<\/li>\n<\/ol>\n\n<!--section=faq-->\n<dl>\n<dt id=\"q%3A%20why%20choose%20axiachat%20ai%20over%20other%20wordpress%20chatbot%20plugins%3F\"><h3>Q: Why choose AxiaChat AI over other WordPress chatbot plugins?<\/h3><\/dt>\n<dd><p>A: AxiaChat AI gives you complete control with support for multiple AI providers (OpenAI, Claude, Gemini), advanced RAG capabilities, MCP server integration, and enterprise features\u2014all while keeping your data on your server. Think of it as your own private ChatGPT, trained on your content, embedded in your site, with no monthly SaaS fees beyond your chosen AI provider's API costs.<\/p><\/dd>\n<dt id=\"q%3A%20do%20i%20need%20an%20openai%20api%20key%3F\"><h3>Q: Do I need an OpenAI API key?<\/h3><\/dt>\n<dd><p>A: You need at least one AI provider API key (OpenAI, Claude, or Gemini). OpenAI is recommended for getting started. All keys are yours\u2014we never charge markup or routing fees.<\/p><\/dd>\n<dt id=\"q%3A%20how%20much%20does%20it%20cost%20to%20run%3F\"><h3>Q: How much does it cost to run?<\/h3><\/dt>\n<dd><p>A: The plugin is free. You only pay your AI provider's API usage (typically $0.001-0.01 per conversation). Usage limits help you control costs. A typical e-commerce site handles hundreds of conversations for under $10\/month.<\/p><\/dd>\n<dt id=\"q%3A%20can%20the%20chatbot%20learn%20from%20my%20website%20content%3F\"><h3>Q: Can the chatbot learn from my website content?<\/h3><\/dt>\n<dd><p>A: Yes! The Easy Config Wizard automatically scans and indexes your site content. You can also upload PDFs, select specific pages, or train it on your knowledge base. The chatbot uses Retrieval Augmented Generation (RAG) to provide accurate, contextual answers.<\/p><\/dd>\n<dt id=\"q%3A%20will%20this%20work%20with%20my%20woocommerce%20store%3F\"><h3>Q: Will this work with my WooCommerce store?<\/h3><\/dt>\n<dd><p>A: Absolutely. The wizard auto-detects WooCommerce and indexes your products. The chatbot can answer product questions, help with comparisons, and guide customers through your catalog.<\/p><\/dd>\n<dt id=\"q%3A%20how%20do%20i%20control%20api%20costs%3F\"><h3>Q: How do I control API costs?<\/h3><\/dt>\n<dd><p>A: Set daily usage limits (per-user, per-IP, or global). Choose what happens when limits are reached: disable input or hide the widget. Monitor usage in real-time via the built-in logs and cost tracking.<\/p><\/dd>\n<dt id=\"q%3A%20is%20chat%20conversation%20data%20stored%3F%20can%20i%20disable%20logging%3F\"><h3>Q: Is chat conversation data stored? Can I disable logging?<\/h3><\/dt>\n<dd><p>A: Chat conversation logging is optional and fully controllable. Toggle it ON to review interactions and improve your bot, or OFF for maximum privacy. When enabled, data stays on YOUR server\u2014never sent to us.<\/p><\/dd>\n<dt id=\"q%3A%20how%20is%20gdpr%20compliance%20handled%3F\"><h3>Q: How is GDPR compliance handled?<\/h3><\/dt>\n<dd><p>A: Built-in GDPR consent management with an in-stream consent bubble that blocks chat input until accepted. IP addresses are only stored if usage limits are enabled. You maintain full data control and can delete logs anytime.<\/p><\/dd>\n<dt id=\"q%3A%20what%20is%20mcp%20server%20integration%3F\"><h3>Q: What is MCP Server Integration?<\/h3><\/dt>\n<dd><p>A: Model Context Protocol (MCP) lets your bot connect to external tools, databases, and APIs. Extend functionality beyond Q&amp;A: fetch live data, integrate with your CRM, query inventory systems, and more\u2014all without custom coding.<\/p><\/dd>\n<dt id=\"q%3A%20does%20it%20support%20web%20search%3F\"><h3>Q: Does it support web search?<\/h3><\/dt>\n<dd><p>A: Yes! Enable provider-native web search so your bot can access current information from the internet. Works with OpenAI GPT-5*, Claude 4.x, and Gemini. Optional domain allowlists let you restrict sources to trusted sites only.<\/p><\/dd>\n<dt id=\"q%3A%20can%20i%20customize%20the%20appearance%3F\"><h3>Q: Can I customize the appearance?<\/h3><\/dt>\n<dd><p>A: Complete customization: colors, avatars (9 included + custom upload), position, size, window controls (draggable, minimizable, closable), button labels, and more. Match your brand perfectly.<\/p><\/dd>\n<dt id=\"q%3A%20where%20can%20i%20deploy%20the%20chatbot%3F\"><h3>Q: Where can I deploy the chatbot?<\/h3><\/dt>\n<dd><p>A: Two ways: (1) Floating global widget visible across your entire site, or (2) inline embedding via shortcode on specific pages\/posts. Mix and match different bots for different pages.<\/p><\/dd>\n<dt id=\"q%3A%20what%20languages%20are%20supported%3F\"><h3>Q: What languages are supported?<\/h3><\/dt>\n<dd><p>A: The AI models support 100+ languages naturally. The plugin UI is translation-ready (Spanish included) and works with RTL languages. Customize all bot responses for any language.<\/p><\/dd>\n<dt id=\"q%3A%20can%20visitors%20send%20images%20or%20documents%20in%20the%20chat%3F\"><h3>Q: Can visitors send images or documents in the chat?<\/h3><\/dt>\n<dd><p>A: Yes! When file uploads are enabled on a bot, visitors see a \ud83d\udcce button to attach photos (JPG, PNG, WebP) or PDF documents. Images are analyzed by the AI using vision capabilities \u2014 the model literally sees the photo. PDFs have their text extracted and analyzed in context. Great for getting quotes from photos, reviewing contracts, diagnosing issues from screenshots, and much more.<\/p><\/dd>\n<dt id=\"q%3A%20can%20the%20ai%20forward%20uploaded%20images%20to%20external%20systems%3F\"><h3>Q: Can the AI forward uploaded images to external systems?<\/h3><\/dt>\n<dd><p>A: Yes. Uploaded images are stored securely as base64 (never exposed via public URLs). When the AI calls a Custom Webhook or MCP Server tool, it can include the image data as a parameter \u2014 making it available to automation platforms like n8n, Zapier, or Make for further processing. Your visitors' private files are never publicly accessible.<\/p><\/dd>\n<dt id=\"q%3A%20can%20visitors%20use%20voice%20input%3F\"><h3>Q: Can visitors use voice input?<\/h3><\/dt>\n<dd><p>A: Yes, voice chat input is supported for compatible chatbots. Users can speak their questions instead of typing.<\/p><\/dd>\n<dt id=\"q%3A%20will%20more%20ai%20providers%20be%20added%3F\"><h3>Q: Will more AI providers be added?<\/h3><\/dt>\n<dd><p>A: Yes! Azure OpenAI, additional models, and more providers are on the roadmap. The plugin's extensible architecture makes adding new providers straightforward.<\/p><\/dd>\n<dt id=\"q%3A%20what%20if%20i%20need%20a%20custom%20capability%20not%20listed%3F\"><h3>Q: What if I need a custom capability not listed?<\/h3><\/dt>\n<dd><p>A: We can build custom capabilities tailored to your business needs. Custom development is available under a project quote. <a href=\"https:\/\/axiachat.org\/\">Contact us for a quote \u2192<\/a><\/p><\/dd>\n<dt id=\"q%3A%20are%20there%20paid%20plans%3F%20what%20do%20they%20include%3F\"><h3>Q: Are there paid plans? What do they include?<\/h3><\/dt>\n<dd><p>A: Yes! AxiaChat AI offers a generous free version with core features. Premium plans unlock additional capabilities. Visit <a href=\"https:\/\/axiachat.org\/\">axiachat.org<\/a> to learn more.<\/p><\/dd>\n<dt id=\"q%3A%20is%20support%20available%3F\"><h3>Q: Is support available?<\/h3><\/dt>\n<dd><p>A: Free users can get community help via the WordPress.org support forum. Priority support with faster response times is available for paid plan subscribers. Visit <a href=\"https:\/\/axiachat.org\/\">axiachat.org<\/a> for support options. When reporting issues, please include your WP version, PHP version, and logging status. You can also export a diagnostics ZIP from the plugin's Advanced settings to speed up troubleshooting.<\/p><\/dd>\n\n<\/dl>\n\n<!--section=changelog-->\n<h4>4.1.1<\/h4>\n\n<ul>\n<li>Added: OpenAI GPT-5.4 family \u2014 <code>gpt-5.4<\/code> (1M context, $2.50\/$15), <code>gpt-5.4-mini<\/code> (400K, $0.75\/$4.50), and <code>gpt-5.4-nano<\/code> (400K, $0.20\/$1.25). GPT-5.4 is now the default model for new bots.<\/li>\n<li>Added: Google Gemini 3.1 models \u2014 <code>gemini-3.1-pro-preview<\/code> ($2\/$12) and <code>gemini-3.1-flash-lite-preview<\/code> ($0.25\/$1.50).<\/li>\n<li>Updated: Gemini model pricing aligned with current Google API rates (2.5 Pro, 2.5 Flash, 2.5 Flash-Lite, 3 Flash).<\/li>\n<li>Updated: Gemini 3 Pro Preview marked as shut down (deprecated Mar 9 2026) \u2014 use Gemini 3.1 Pro Preview.<\/li>\n<li>Updated: Gemini 2.0 Flash \/ Flash-Lite marked as legacy (deprecated Jun 2026).<\/li>\n<\/ul>\n\n<h4>3.1.7<\/h4>\n\n<ul>\n<li>Changed: Footer attribution is now opt-in for all users (default off). Site administrators can enable it from Settings \u2192 Add-ons or during the Easy Setup Wizard.<\/li>\n<li>Improved: External Services documentation updated \u2014 added Google Gemini and Freemius SDK disclosures.<\/li>\n<\/ul>\n\n<h4>3.1.5<\/h4>\n\n<p>\ud83c\udfaf <strong>Lead Capture \u2014 Major Overhaul<\/strong>\n* Added: Inline lead forms (<code>show_form<\/code>) \u2014 the bot displays a styled form directly inside the chat widget. Visitors fill it out and submit without leaving the conversation.\n* Added: Multiple Lead Lists \u2014 create independent lists, each with its own fields, destination, form design, bot assignment, and notifications.\n* Added: Form Appearance customization per list \u2014 display mode (full\/compact), custom HTML header, submit button text, success message, background color, and button color.\n* Added: Live form preview panel in admin \u2014 see real-time changes to form appearance, fields, types, and required markers as you edit.\n* Added: Custom field types \u2014 Text, Email, Phone, Textarea, Number, and URL with proper validation and mobile keyboard support.\n* Added: Required field support \u2014 mark fields as mandatory with red asterisk indicator and client-side validation.\n* Added: Bot assignment per list \u2014 assign a list to all bots or restrict it to a specific bot for multi-bot setups.\n* Added: List name column in leads table and list filter for multi-list management.\n* Added: Total leads footer row in lists table with global count.\n* Improved: <code>show_form<\/code> is now the primary recommended capture method; <code>save_lead<\/code> (conversational) remains available as fallback for non-visual channels (WhatsApp, SMS).\n* Improved: Chat input is temporarily disabled while the inline form is visible to focus visitor attention.\n* Improved: Form submit shows a customizable success message with green checkmark, then conversation resumes normally.\n* Fixed: Lead list filter and per-list count now work correctly (list_id properly stored on capture).<\/p>\n\n<h4>3.1.0<\/h4>\n\n<ul>\n<li>Added: Quick Questions \u2014 predefined question buttons displayed above the chat input to guide visitors and boost engagement from the first interaction.<\/li>\n<li>Added: Per-bot Quick Questions configuration \u2014 enable\/disable and manage questions from Bots \u2192 Extras with dynamic add\/remove fields and auto-save.<\/li>\n<\/ul>\n\n<h4>3.0.9<\/h4>\n\n<ul>\n<li>Added: File &amp; Image Uploads \u2014 visitors can now send photos (JPG, PNG, WebP) and PDF documents directly in the chat widget via a new \ud83d\udcce attachment button.<\/li>\n<li>Added: AI Vision support \u2014 uploaded images are sent to the AI model's vision capability (GPT-4o, Claude Sonnet\/Opus, Gemini Pro\/Flash) for real-time visual analysis.<\/li>\n<li>Added: PDF text extraction \u2014 uploaded PDFs have their text extracted automatically and injected into the conversation context.<\/li>\n<li>Added: Automatic image optimization \u2014 uploaded images are resized (max 1536px) and compressed to reduce API costs and speed up processing.<\/li>\n<li>Added: In-chat file preview \u2014 sent messages show a thumbnail (images) or file indicator (PDFs) attached to the user's message bubble.<\/li>\n<li>Added: Admin log file indicators \u2014 conversation log list shows \ud83d\udcce icon for sessions with attachments; detail view shows clickable thumbnails with secure full-size preview (base64 data URI, no public file exposure).<\/li>\n<li>Added: Per-bot upload configuration \u2014 enable\/disable uploads, select allowed file types (PDF, JPG, PNG, WebP), and set max file size (1\u201320 MB) independently for each bot.<\/li>\n<li>Security: Uploaded files are never exposed via public URLs \u2014 images are stored as base64 in the database, accessible only to authenticated administrators.<\/li>\n<\/ul>\n\n<h4>3.0.8<\/h4>\n\n<ul>\n<li>Added: OpenAI GPT-5.3 model support (<code>gpt-5.3-chat-latest<\/code>) \u2014 now the default model for new bots.<\/li>\n<li>Improved: Full page-cache compatibility \u2014 automatic cache purge when bot settings change, supporting WP Rocket, LiteSpeed Cache, W3 Total Cache, WP Super Cache, SiteGround Optimizer, WP Fastest Cache, WP-Optimize, Cache Enabler, Hummingbird, Breeze (Cloudways), Kinsta, Cloudflare Super Page Cache, Autoptimize, and Nginx Helper. Third-party plugins can hook into <code>aichat_after_purge_cache<\/code>.<\/li>\n<li>Added: View and Edit buttons for indexed documents in Training \u2192 Context \u2014 inspect individual chunks, spot indexing errors, and surgically edit specific chunks with automatic re-embedding on save.<\/li>\n<\/ul>\n\n<h4>3.0.7<\/h4>\n\n<ul>\n<li>Added: GDPR Consent now supports two modes \u2014 classic accept button or an email form gate that captures visitor contact details before the chat starts.<\/li>\n<li>Added: Email form mode collects email (required), name, and phone (both optional toggles with configurable placeholders).<\/li>\n<li>Added: GDPR Contacts tab under Leads \u2014 view, search, delete, bulk-delete, and export contacts captured via the email form. No plan limits apply.<\/li>\n<li>Added: \"Delete all plugin data on uninstall\" option in Settings \u2192 Advanced \u2014 when enabled, uninstalling the plugin removes all database tables and settings; disabled by default to prevent accidental data loss.<\/li>\n<\/ul>\n\n<h4>3.0.5<\/h4>\n\n<ul>\n<li>Added: WhatsApp CTA button \u2014 optionally show a WhatsApp contact button inside the chat window, configurable per bot (phone number, custom message, button label, and display position).<\/li>\n<li>Added: Automatic front-end translation <\/li>\n<\/ul>\n\n<h4>3.0.4<\/h4>\n\n<ul>\n<li>Changed: Google Gemini is now the default and first provider in the Easy Config wizard \u2014 free tier covers ~100 messages\/day.<\/li>\n<li>Added: \"How to get a free API key\" tip banner below provider cards in Easy Config (step 3) with step-by-step modal guide.<\/li>\n<li>Added: Same free API key tip banner on the Settings \u2192 General page when no API keys are configured.<\/li>\n<li>Improved: Provider card order updated across the plugin: Google Gemini \u2192 OpenAI \u2192 Anthropic.<\/li>\n<\/ul>\n\n<h4>3.0.3<\/h4>\n\n<ul>\n<li>Added: Email alerts for new conversations \u2014 receive an email each time a user finishes chatting with your bot (enabled by default, configurable under Settings \u2192 General).<\/li>\n<li>Added: Advanced email alert options modal \u2014 choose between full transcript or summary-only emails, per-conversation or digest mode, and configurable idle timeout.<\/li>\n<\/ul>\n\n<h4>3.0.2<\/h4>\n\n<ul>\n<li>Fixed: Tool continuation no longer triggers embedding generation (prevented 400 errors with Gemini provider).<\/li>\n<li>Fixed: Tool execution continuation now works reliably for Gemini and Claude providers (silent death resolved).<\/li>\n<li>Fixed: Suggestions chips now render correctly after tool continuation responses.<\/li>\n<li>Fixed: GPT-5.x hallucinated tool narration (e.g. \"(executing tools\u2026)\") is detected and retried automatically.<\/li>\n<\/ul>\n\n<h4>3.0.1<\/h4>\n\n<p>\ud83c\udfa8 <strong>Redesign &amp; User-Friendly Overhaul<\/strong>\n* Redesigned: Brand-new Easy Config wizard \u2014 set up a fully working bot in under 2 minutes with a guided, step-by-step flow.\n* Redesigned: Training \u2192 Instructions now uses a clean card-based layout with built-in universal guidelines and a \"Quick mode \/ Full editor\" toggle.\n* Redesigned: Training \u2192 Context page rebuilt with a modern batch-processing UI, real-time progress, and clearer status indicators.\n* Redesigned: Bot Settings reorganized into logical tabs (General, Appearance, Window, Advanced) with live preview and simplified controls.\n* Added: One-click AI-generated bot instructions \u2014 describe what you need and the AI writes the full system prompt for you.\n* Added: Built-in prompt templates (Customer Support, Lead Generation, E-commerce Assistant, etc.) selectable from the Instructions page.\n* Improved: AI Tools settings simplified \u2014 per-tool System Policy removed; all bot behavior is now centralized in Training \u2192 Instructions with clear usage tips and examples.\n* Improved: Full JavaScript internationalization audit \u2014 all admin UI strings are now translatable.<\/p>\n\n<h4>2.0.9<\/h4>\n\n<ul>\n<li>Added: Advanced Indexing Options per context \u2014 configure which fields are embedded (excerpt, URL, featured image, taxonomies, custom meta keys).<\/li>\n<li>Added: WooCommerce static fields in indexing \u2014 short description and attributes can now be included in embeddings.<\/li>\n<li>Added: WooCommerce dynamic enrichment \u2014 price, stock status and SKU are injected at query time with fresh data (no re-indexing needed when prices change).<\/li>\n<li>Added: Featured image URL option \u2014 when enabled, the AI can show product\/post images directly in the chat using Markdown.<\/li>\n<li>Added: Image rendering support in chat widget \u2014 bot responses can now display images via Markdown syntax.<\/li>\n<li>Added: Post URL included by default in embeddings \u2014 the AI can now reference direct links to content. Configurable in Advanced Indexing Options.<\/li>\n<li>Fixed: Pinecone upsert metadata now includes content and post type \u2014 fixes empty context and enables WooCommerce enrichment for Pinecone-based contexts.<\/li>\n<li>Improved: Markdown-to-HTML fallback now supports image syntax <code>![alt](url)<\/code>.<\/li>\n<\/ul>\n\n<h4>2.0.8<\/h4>\n\n<ul>\n<li>Added: Knowledge Sources Limit setting (3\u201310 slider) in Thresholds \u2014 control how many RAG chunks are included in each AI response (default 5).<\/li>\n<li>Added: History Persistence toggle per bot \u2014 when OFF, chat history is not restored on page refresh.<\/li>\n<li>Improved: Context Max Length default raised from 4096 to 6000 characters to avoid truncating rich RAG contexts.<\/li>\n<\/ul>\n\n<h4>2.0.7<\/h4>\n\n<ul>\n<li>Contexts can now be modified.<\/li>\n<li>Preview page to test the bot before deploying it globally.<\/li>\n<li>Button in the widget to clear conversation history.<\/li>\n<\/ul>\n\n<h4>2.0.6<\/h4>\n\n<p>\ud83d\udee1\ufe0f <strong>CSS Isolation \u2013 Theme-Proof Styling<\/strong>\n* Fixed: Chat widget text and background colors are now hardened with <code>!important<\/code> declarations to prevent WordPress themes from overriding them.\n* Fixed: Dark themes can no longer cause invisible text (white-on-white) inside chat bubbles.\n* Hardened: All critical UI elements protected \u2014 message bubbles, input field, send button, header, suggestions chips, GDPR consent block, footer, and super-minimized avatar.<\/p>\n\n<p>\ud83d\udd27 <strong>Diagnostics ZIP Export<\/strong>\n* Added: One-click \"Export Diagnostics ZIP\" button in Settings \u2192 Advanced tab for faster remote support.\n* Includes: Full environment report (WP\/PHP\/MySQL versions, active theme, active plugins, bot &amp; context summaries) plus plugin log files. Never includes API keys or secrets.<\/p>\n\n<h4>2.0.2<\/h4>\n\n<p><strong>Major Release: Business Automation Suite<\/strong><\/p>\n\n<p>\ud83c\udfaf <strong>Lead Capture System<\/strong><\/p>\n\n<p>\ud83d\udcc5 <strong>Appointment Booking System<\/strong><\/p>\n\n<p>\ud83d\udee0\ufe0f <strong>Other Improvements<\/strong>\n* Updated: Default model changed to GPT-5.2-chat-latest (December 2025).\n* Improved: Provider registry architecture for easier extensibility.\n* Improved: Tool execution with better error handling and logging.\n* Fixed: Easy Config wizard now uses GPT-5.2-chat-latest as default model.\n* Fixed: Model selection properly passed through Agency proxy.<\/p>\n\n<h4>1.2.9<\/h4>\n\n<ul>\n<li>Added: Markdown rendering support for AI responses (safe HTML output).<\/li>\n<li>Added: Optional \u201cSuggestions \/ Next actions\u201d chips in the widget (configurable count and colors).<\/li>\n<li>Improved: Usage logging accuracy for OpenAI Responses API (GPT-5.x \/ GPT-5.2) and cost calculation.<\/li>\n<li>Improved: Usage screen with \u201cLast Conversations\u201d view for quick auditing.<\/li>\n<li>Fixed: Preview home mode now avoids inline CSS\/JS injection (enqueue-based).<\/li>\n<\/ul>\n\n<h4>1.2.8<\/h4>\n\n<ul>\n<li>Added: GPT-5.2 models support (Instant, Thinking) - OpenAI's December 2025 generation.<\/li>\n<li>Updated: Token limits and pricing for new GPT-5.2 models.<\/li>\n<\/ul>\n\n<h4>1.2.7<\/h4>\n\n<ul>\n<li>Added: GPT-5.1 models support (Instant, Thinking) - OpenAI's latest generation.<\/li>\n<li>Tested: Fully compatible with WordPress 6.9.<\/li>\n<li>Updated: Token limits and pricing for new GPT-5.1 models.<\/li>\n<\/ul>\n\n<h4>1.2.6<\/h4>\n\n<ul>\n<li>Added: WhatsApp as an input\/output channel through integration with another plugin.<\/li>\n<li>Enhanced: Tabbed interface in settings for better navigation.<\/li>\n<li>Added: New \"Advanced\" tab in settings.<\/li>\n<\/ul>\n\n<h4>1.2.5<\/h4>\n\n<ul>\n<li>Added: Support for Google Gemini as a new AI provider.<\/li>\n<\/ul>\n\n<h4>1.2.4<\/h4>\n\n<ul>\n<li>Improved: Encryption of AI model API keys for better security.<\/li>\n<\/ul>\n\n<h4>1.2.3<\/h4>\n\n<ul>\n<li>Added: YouTube tutorial video for quick setup walkthrough<\/li>\n<li>Enhanced: Flexible widget sizing for better responsive design<\/li>\n<li><p>Improved: Larger avatar display for better visual prominence<\/p><\/li>\n<li><p>GDPR: Redesigned the consent message as a full-width in-stream block with a darker background and a centered button for better visibility and UX. Inputs remain disabled until accepted.<\/p><\/li>\n<\/ul>\n\n<h4>1.2.2<\/h4>\n\n<ul>\n<li>Admin: Improved Bots accordion behavior \u2014 clicking an open section now closes it (allowing zero-open state) while keeping \u201cat most one open\u201d when opening another.<\/li>\n<li>Admin: Reduced console noise; detailed accordion debug logs are now disabled by default and can be re-enabled with ?aichat_debug=1.<\/li>\n<li>Preview: Stabilized homepage preview and suppressed duplicate global widget when previewing a specific bot.<\/li>\n<\/ul>\n\n<!--section=description-->\n<p>The best AI Chatbot for WordPress. Like having ChatGPT trained on your content \u2014 turn your site into a 24\/7 sales &amp; support machine.<\/p>\n\n<h3>\ud83e\udd16 The best free AI Chatbot for WordPress<\/h3>\n\n<p>If you want more leads, more sales, and less repetitive support work, you need a powerful chatbot on your website.<\/p>\n\n<p>AxiaChat AI is the best AI chatbot for WordPress designed to help you automate conversations, capture opportunities, and assist visitors instantly \u2014 24 hours a day.<\/p>\n\n<p>This is not just another basic chatbot.<br \/>\nThis is a smart, business-ready chatbot built for real results \u2014 like having a private ChatGPT that knows your business inside out.<\/p>\n\n<h3>\u26a1 Why This Chatbot Is Different<\/h3>\n\n<p>Most WordPress chatbot plugins only provide simple scripted replies.<\/p>\n\n<p>AxiaChat AI gives you a real AI-powered chatbot \u2014 backed by the same technology as ChatGPT \u2014 that understands context, learns from your content, and responds like a human assistant.<\/p>\n\n<p>Your chatbot can:<\/p>\n\n<p>\u2705 <strong>Answer product questions<\/strong> automatically, 24\/7\n\u2705 <strong>Guide customers<\/strong> to the right service or product\n\u2705 <strong>Capture leads<\/strong> automatically \u2014 even when you're offline\n\u2705 <strong>Book appointments<\/strong> directly from the chat\n\u2705 <strong>Analyze images &amp; documents<\/strong> \u2014 visitors send photos or PDFs, the AI sees and responds\n\u2705 <strong>Reduce support workload<\/strong> by handling repetitive questions\n\u2705 <strong>Work 24\/7<\/strong> without breaks, staff costs, or delays<\/p>\n\n<p>This chatbot becomes part of your team. That's what real AI chat for WordPress looks like.<\/p>\n\n<h3>\u2b50 What Users Say<\/h3>\n\n<p>\u2b50\u2b50\u2b50\u2b50\u2b50 <strong>\"The only WooCommerce chatbot plugin that actually works\"<\/strong>\n\"Congratulations to the developer for an excellent job. The plugin is well built, reliable, and delivers exactly what it promises.\" \u2014 gabricp100, February 2026<\/p>\n\n<p>\u2b50\u2b50\u2b50\u2b50\u2b50 <strong>\"My conversion rate went up thanks to this chatbot!\"<\/strong>\n\"Perfect for small businesses. My conversion rate went up thanks to this chatbot!\" \u2014 david25k, February 2026<\/p>\n\n<p>\u2b50\u2b50\u2b50\u2b50\u2b50 <strong>\"Customizable and powerful\"<\/strong>\n\"I love how customizable it is, you can train your own AI agent and make it fit your brand.\" \u2014 queroamilys, March 2026<\/p>\n\n<p>\u2b50\u2b50\u2b50\u2b50\u2b50 <strong>\"It works great!\"<\/strong>\n\"We implemented it in the store and it works great.\" \u2014 Antonio Aponte, February 2026<\/p>\n\n<p><strong>12 reviews on WordPress.org \u00b7 100% five-star ratings.<\/strong>\n<a href=\"https:\/\/wordpress.org\/support\/plugin\/axiachat-ai\/reviews\/\">Read all reviews \u2192<\/a><\/p>\n\n<h3>\ud83d\udcc8 Built for Growth<\/h3>\n\n<p>Whether you run an eCommerce store, agency, SaaS, or local business, this chatbot helps you scale without hiring more staff.<\/p>\n\n<p>Instead of answering the same questions every day, let your chatbot handle them.<\/p>\n\n<p>Instead of missing leads after hours, let your chatbot capture them.<\/p>\n\n<p>Instead of losing visitors, let your chatbot engage them instantly.<\/p>\n\n<p>A smart chatbot increases conversions.<br \/>\nA fast chatbot improves user experience.<br \/>\nA trained chatbot builds trust.<\/p>\n\n<h3>\ud83c\udf93 Train Your Chatbot on Your Own Content<\/h3>\n\n<p>What makes this chatbot powerful is its ability to learn from your website.<\/p>\n\n<p>You can train your chatbot using:<\/p>\n\n<p>\ud83d\udcd8 Pages and posts\n\ud83d\uded2 WooCommerce products (prices, stock, SKU \u2014 always up to date)\n\u2753 FAQs\n\ud83d\udcc4 PDFs\n\ud83e\udde0 Custom knowledge bases\n\ud83c\udf10 External web pages<\/p>\n\n<p>And your visitors can send files directly in the chat:<\/p>\n\n<p>\ud83d\udcf8 <strong>Photos &amp; screenshots<\/strong> \u2014 the AI literally sees the image using vision AI\n\ud83d\udcc4 <strong>PDF documents<\/strong> \u2014 text is extracted automatically and analyzed in context<\/p>\n\n<p>Imagine a customer sending a photo of their broken appliance, or uploading a contract for the AI to summarize. That's not future tech \u2014 it works today.<\/p>\n\n<p>This means your chatbot gives accurate answers based on YOUR business \u2014 not generic internet responses.<\/p>\n\n<p>Your visitors talk to a specialized chatbot, not a random AI. Genuine AI chat powered by your own knowledge base.<\/p>\n\n<h3>\ud83c\udfaf Designed for Conversions<\/h3>\n\n<p>This is not just a support chatbot.<\/p>\n\n<p>It is also a sales chatbot.<\/p>\n\n<p>Your chatbot can:<\/p>\n\n<p>\ud83c\udfaf Collect names and emails\n\ud83d\udce7 Send notifications to your team\n\ud83d\udcc5 Book appointments\n\ud83d\udc64 Escalate to a human agent\n\ufffd Receive photos &amp; documents from visitors and act on them\n\ud83d\udd17 Trigger custom workflows including forwarding uploaded images securely<\/p>\n\n<p>Every AI chat conversation with your chatbot becomes an opportunity.<\/p>\n\n<h3>\u26a1 Easy Setup in Minutes<\/h3>\n\n<p>You don't need technical skills to launch your chatbot.<\/p>\n\n<ol>\n<li>Install the plugin<\/li>\n<li>Add your API key<\/li>\n<li>Run the setup wizard<\/li>\n<li>Activate your chatbot<\/li>\n<\/ol>\n\n<p>In less than 5 minutes, your AI chatbot is live and your AI chat channel is open for visitors.<\/p>\n\n<h3>\ud83c\udfa8 Fully Customizable Chatbot<\/h3>\n\n<p>Match your brand perfectly.<\/p>\n\n<p>Your chatbot supports:<\/p>\n\n<p>\ud83c\udfa8 Custom colors and footer\n\ud83e\udd16 Avatars (9 included + custom upload)\n\ufffd File upload button (images &amp; PDFs) with in-chat preview\n\ufffd\ud83d\udcf1 Floating widget or inline embed via shortcode\n\u270d\ufe0f Draggable, minimizable, and closable window\n\ud83d\udcf2 Fully mobile responsive design<\/p>\n\n<p>Make your chatbot look like it was built specifically for your site.<\/p>\n\n<h3>\ud83d\udcb0 Control Costs &amp; Usage<\/h3>\n\n<p>Worried about AI costs?<\/p>\n\n<p>Your chatbot includes:<\/p>\n\n<p>\ud83d\udd22 Daily usage limits (global and per-user)\n\ud83d\udeab Per-IP and per-user restrictions\n\ud83d\udcca Real-time cost tracking and token usage\n\ud83d\udcdd Optional conversation logging (on\/off anytime)<\/p>\n\n<p>You stay in control while your chatbot works for you.<\/p>\n\n<h3>\ud83e\udde0 Powerful AI Behind Your Chatbot<\/h3>\n\n<p>Your chatbot can connect to:<\/p>\n\n<p>\ud83d\udfe2 <strong>OpenAI<\/strong> \u2014 GPT-5.4, GPT-5.4 Mini, GPT-5.3, GPT-5, GPT-4.1 and all current models\n\ud83d\udfe2 <strong>Anthropic Claude<\/strong> \u2014 Opus 4.6, Sonnet 4.5, Haiku 4.5 and more\n\ud83d\udfe2 <strong>Google Gemini<\/strong> \u2014 3.1 Pro, 3 Flash, 2.5 Flash, 2.5 Pro (free tier available)<\/p>\n\n<p>You choose the AI engine for your chatbot.<br \/>\nYou control how your chatbot behaves.<br \/>\nYou decide what your chatbot knows.<\/p>\n\n<p>Unlike SaaS platforms or using ChatGPT directly, this chatbot runs directly on your WordPress site.<br \/>\nYour data stays on your server. Your brand stays front and center. Your AI chat, your rules.<\/p>\n\n<h3>\ud83d\udd10 GDPR &amp; Privacy Friendly<\/h3>\n\n<p>This chatbot includes built-in consent tools and privacy controls.<\/p>\n\n<p>You can enable or disable conversation logging anytime.<\/p>\n\n<p>Your chatbot respects your users and your compliance requirements.<\/p>\n\n<h3>\u2705 Perfect For<\/h3>\n\n<p>\ud83d\uded2 WooCommerce stores\n\ud83c\udfe2 Agencies\n\ud83d\ude80 SaaS businesses\n\ud83c\udfe4 Corporate websites\n\ud83d\udd27 Service companies\n\u270d\ufe0f Bloggers\n\ud83c\udf93 Online educators<\/p>\n\n<p>If you want more engagement and automation, you need a professional chatbot and a reliable AI chat solution.<\/p>\n\n<h3>\ud83d\udcda Documentation &amp; Technical Resources<\/h3>\n\n<p>Official website:\n<a href=\"https:\/\/axiachat.org\/\">axiachat.org<\/a><\/p>\n\n<p>Full documentation, tutorials and advanced configuration guides:\n<a href=\"https:\/\/axiachat.org\/help\/\">axiachat.org\/help<\/a><\/p>\n\n<p>Community support forum:\n<a href=\"https:\/\/wordpress.org\/support\/plugin\/axiachat-ai\/\">wordpress.org\/support\/plugin\/axiachat-ai<\/a><\/p>\n\n<h3>\ud83c\udd93 Getting Started<\/h3>\n\n<p>The free version of this chatbot is fully functional and covers most business needs.<\/p>\n\n<p>Start for free, explore all the core features, and grow at your own pace.<\/p>\n\n<h3>\ud83d\ude80 Stop Losing Opportunities<\/h3>\n\n<p>Every visitor that leaves your site without interacting is a missed opportunity.<\/p>\n\n<p>A smart chatbot greets them.<br \/>\nA helpful chatbot answers them.<br \/>\nA proactive chatbot converts them.<\/p>\n\n<p>Install the best AI chatbot for WordPress today \u2014 your own ChatGPT working exclusively for your business \u2014 and transform your website into a 24\/7 growth engine.<\/p>\n\n<h3>Technical Features<\/h3>\n\n<h4>Supported AI Models<\/h4>\n\n<p><strong>OpenAI (GPT)<\/strong>\n* GPT-5.4 \u2014 <code>gpt-5.4<\/code> \u2014 1M context, thinking + multimodal (recommended, new)\n* GPT-5.4 Mini \u2014 <code>gpt-5.4-mini<\/code> \u2014 400K context, thinking + multimodal (new)\n* GPT-5.4 Nano \u2014 <code>gpt-5.4-nano<\/code> \u2014 400K context, ultra-low cost $0.20\/$1.25 per 1M tokens (new)\n* GPT-5.3 Instant \u2014 <code>gpt-5.3-chat-latest<\/code> \u2014 256K context, multimodal\n* GPT-5.2 Instant \u2014 <code>gpt-5.2-chat-latest<\/code> \u2014 256K context, multimodal\n* GPT-5.2 Thinking \u2014 <code>gpt-5.2<\/code> \u2014 adaptive reasoning mode\n* GPT-5 \u2014 <code>gpt-5<\/code> \u2014 256K context, thinking + multimodal\n* GPT-5 Mini \u2014 <code>gpt-5-mini<\/code> \u2014 128K context, thinking\n* GPT-5 Nano \u2014 <code>gpt-5-nano<\/code> \u2014 64K context, ultra-low cost ($0.05\/$0.40 per 1M tokens)\n* GPT-4.1 \/ GPT-4.1 Mini \/ GPT-4.1 Nano \u2014 up to 1M context window\n* GPT-4o \/ GPT-4o Mini \u2014 multimodal, vision capable\n* GPT-4 Turbo \u2014 legacy, 128K context\n* GPT-3.5 Turbo \u2014 legacy, cost-effective<\/p>\n\n<p><strong>Anthropic (Claude)<\/strong>\n* Claude Opus 4.6 \u2014 <code>claude-opus-4-6<\/code> \u2014 200K context, thinking (Feb 2026)\n* Claude Sonnet 4.5 \u2014 <code>claude-sonnet-4-5<\/code> \u2014 1M context, thinking (recommended)\n* Claude Haiku 4.5 \u2014 <code>claude-haiku-4-5<\/code> \u2014 1M context, fastest\n* Claude Opus 4.5 \/ Opus 4.1 \/ Opus 4 \u2014 200K context, advanced reasoning\n* Claude Sonnet 4 \u2014 200K context, thinking\n* Claude Sonnet 3.7 \/ Haiku 3.5 \/ Haiku 3 \u2014 legacy<\/p>\n\n<p><strong>Google Gemini<\/strong>\n* Gemini 3.1 Pro Preview \u2014 <code>gemini-3.1-pro-preview<\/code> \u2014 1M context, thinking (new)\n* Gemini 3.1 Flash-Lite Preview \u2014 <code>gemini-3.1-flash-lite-preview<\/code> \u2014 1M context, efficient (new)\n* Gemini 3 Flash Preview \u2014 <code>gemini-3-flash-preview<\/code> \u2014 1M context, thinking\n* Gemini 3 Pro Preview \u2014 <code>gemini-3-pro-preview<\/code> \u2014 shut down Mar 2026, use 3.1 Pro instead\n* Gemini 2.5 Pro \u2014 <code>gemini-2.5-pro<\/code> \u2014 1M context, reasoning\n* Gemini 2.5 Flash \u2014 <code>gemini-2.5-flash<\/code> \u2014 1M context, balanced (default, free tier available)\n* Gemini 2.5 Flash-Lite \u2014 <code>gemini-2.5-flash-lite<\/code> \u2014 ultra-efficient ($0.10\/$0.40 per 1M tokens)\n* Gemini 2.0 Flash \/ 2.0 Flash-Lite \u2014 legacy, deprecated Jun 2026<\/p>\n\n<p>All models are managed through a centralised registry (filterable via <code>aichat_model_registry<\/code> hook). New models can be added by third-party add-ons without modifying core files.<\/p>\n\n<h4>Automation &amp; Integrations<\/h4>\n\n<ul>\n<li><strong>REST API ready<\/strong> \u2014 every bot interaction is driven by a WordPress AJAX endpoint, making it straightforward to integrate from external systems<\/li>\n<li><strong>Outbound webhook parameters<\/strong> \u2014 configure method (GET\/POST), headers, body template with dynamic placeholders, and response mapping<\/li>\n<\/ul>\n\n<h4>Add-ons<\/h4>\n\n<ul>\n<li><strong>AI Tools<\/strong> \u2014 native function-calling tools (web search, calculations, date\/time, etc.)<\/li>\n<li><strong>File &amp; Image Uploads<\/strong> \u2014 visitors can send photos (JPG, PNG, WebP) and PDF documents directly in the chat. Images are analyzed via AI vision; PDFs have text extracted automatically. Configurable per bot with file type and size controls<\/li>\n<li><strong>Appointments<\/strong> \u2014 let the bot schedule appointments directly from the chat<\/li>\n<li><strong>Lead Capture<\/strong> \u2014 collect and store visitor data (name, email, phone) inside the chat flow<\/li>\n<li><strong>Web Scraper<\/strong> \u2014 fetch and summarise live web content as part of a bot response<\/li>\n<\/ul>\n\n<h4>RAG \u2014 Retrieval-Augmented Generation<\/h4>\n\n<ul>\n<li>Local vector search using cosine similarity (no external dependency required)<\/li>\n<li><strong>Pinecone<\/strong> vector database support for large-scale knowledge bases<\/li>\n<li>Embeddings via OpenAI <code>text-embedding-3-small<\/code><\/li>\n<li>Advanced indexing options: excerpt, URL, featured image, taxonomies, custom meta keys<\/li>\n<li><strong>WooCommerce enrichment<\/strong> \u2014 price, stock status and SKU injected at query time (no re-indexing needed on price changes)<\/li>\n<li>Featured image URLs included \u2014 the AI can render product images via Markdown in the chat<\/li>\n<\/ul>\n\n<h4>Security &amp; Validation<\/h4>\n\n<ul>\n<li>WordPress nonce (<code>aichat_ajax<\/code>) on every request<\/li>\n<li>Honeypot field \u2014 silent bot protection without CAPTCHA friction<\/li>\n<li>Pluggable CAPTCHA filter (<code>aichat_validate_captcha<\/code>)<\/li>\n<li>OpenAI Moderation API integration (<code>aichat_run_moderation_checks<\/code>)<\/li>\n<li>Per-IP burst rate limiting and spam heuristics<\/li>\n<li>Per-user and global daily usage limits with hard message-length cap (4 000 chars)<\/li>\n<li>System prompt never exposed \u2014 security policy injected first and enforced via refusal instructions<\/li>\n<li>Remote endpoint allowlist for Pinecone and webhook calls<\/li>\n<\/ul>\n\n<h4>Frontend &amp; UI<\/h4>\n\n<ul>\n<li>Floating widget (configurable position: bottom-right, bottom-left, etc.) or inline layout via shortcode<\/li>\n<li>Fully draggable, minimizable and closable chat window<\/li>\n<li>Voice chat mode (<code>type=\"voice_text\"<\/code>) with browser Web Speech API<\/li>\n<li>File upload button (\ud83d\udcce) with drag-and-drop support, in-chat thumbnail preview, and admin log attachment indicators<\/li>\n<li>Automatic image optimization (resize + compress) to minimize API costs while preserving visual quality<\/li>\n<li>Markdown rendering including images, bold, lists, and code blocks<\/li>\n<li>Suggestion chips \/ quick-reply buttons<\/li>\n<li>WhatsApp CTA button inside the chat window (configurable per bot)<\/li>\n<li>Multi-bot support \u2014 create multiple bots with independent instructions, models, and context<\/li>\n<li>GDPR consent message injected as first bot message until visitor accepts (<code>aichat_gdpr_ok<\/code> cookie)<\/li>\n<li>Session UUID stored client-side for persistent conversation history across page reloads<\/li>\n<li>Automatic front-end translation \u2014 UI labels localised to the visitor's browser language<\/li>\n<\/ul>\n\n<h4>Admin &amp; Developer Tools<\/h4>\n\n<ul>\n<li>Guided <strong>Easy Setup Wizard<\/strong> \u2014 fully working bot in under 2 minutes<\/li>\n<li>One-click AI-generated bot instructions \u2014 describe your use case, AI writes the system prompt<\/li>\n<li>Built-in prompt templates (Customer Support, Lead Generation, E-commerce, FAQ, etc.)<\/li>\n<li>Chat conversation logs with full transcript, token usage, and cost breakdown<\/li>\n<li>Email alerts on new conversations \u2014 full transcript or summary, per-conversation or digest mode<\/li>\n<li>Debug mode: <code>define('AICHAT_DEBUG', true)<\/code> for server-side verbose logging; <code>?aichat_debug=1<\/code> for JS console<\/li>\n<li>Filterable hooks for system policy (<code>aichat_security_policy<\/code>), model registry, endpoint allowlist, and more<\/li>\n<li><code>[aichat id=\"slug\"]<\/code> shortcode with 20+ data-attribute parameters (layout, position, colours, avatar, voice, etc.)<\/li>\n<\/ul>\n\n<h3>Bundled Libraries<\/h3>\n\n<ul>\n<li>Bootstrap (local)<\/li>\n<li>Bootstrap Icons (local)<\/li>\n<li>smalot\/pdfparser (LGPLv3) for PDF extraction<\/li>\n<\/ul>\n\n<h3>External Services \/ Data Disclosure<\/h3>\n\n<p>This plugin can connect to the following third\u2011party APIs depending on which features you enable. You (the site owner) must supply the API keys. No keys are bundled and no traffic is proxied through a vendor server controlled by this plugin author.<\/p>\n\n<h3>1. OpenAI<\/h3>\n\n<p>Used for: chat completions \/ responses, embeddings (context indexing), moderation (safety checks).<\/p>\n\n<p>Endpoints used (HTTPS):\n* https:\/\/api.openai.com\/v1\/chat\/completions (legacy Chat Completions)\n* https:\/\/api.openai.com\/v1\/responses (Responses API \u2013 new unified endpoint if configured)\n* https:\/\/api.openai.com\/v1\/embeddings (document\/post\/PDF embedding + wizard indexing)\n* https:\/\/api.openai.com\/v1\/moderations (content moderation)<\/p>\n\n<p>Data Sent:\n* User prompt text (per message) and limited conversation history (trimmed for token control)\n* System \/ policy instructions (security + privacy policy + bot instructions)\n* Optional retrieved context snippets (only the selected top\u2011N chunks or page excerpt \u2013 never the full original document)\n* Embedding requests: raw chunk text produced from your site\u2019s content or uploaded PDFs\n* Moderation: only the user prompt text (not the entire history)<\/p>\n\n<p>Data Retention (Your Server):\n* Conversation log rows (if logging enabled) including user prompt, model reply, timestamps, bot slug, session id, optional user id. IP (binary) only stored when per\u2011IP limits are turned on. Disable logging to stop storing new rows.\n* Embeddings table stores numeric vectors generated from your content (not reversible plaintext) plus the original chunk text for retrieval.<\/p>\n\n<p>Recommendations:\n* Update your privacy policy to disclose sending user prompts and limited site content to OpenAI for processing.\n* Disable logging or periodically purge if you process personal data.<\/p>\n\n<p>Legal \/ Docs:\n* Terms: https:\/\/openai.com\/policies\/terms-of-use\n* Privacy: https:\/\/openai.com\/policies\/privacy-policy\n* Usage Policies: https:\/\/openai.com\/policies\/usage-policies<\/p>\n\n<p>Opt\u2011Out \/ Control:\n* Remove the OpenAI API key in Settings to stop all OpenAI calls (bots will refuse to answer).\n* Disable conversation logging.\n* Limit context ingestion to non\u2011sensitive pages.<\/p>\n\n<h3>2. Anthropic (Claude)<\/h3>\n\n<p>Used for: alternative chat completions via Claude models (messages API) when a bot provider is set to Anthropic\/Claude.<\/p>\n\n<p>Endpoint used (HTTPS):\n* https:\/\/api.anthropic.com\/v1\/messages<\/p>\n\n<p>Data Sent:\n* A rewritten message array: system instructions + user prompt + condensed prior turns (trimmed) + optional retrieved context snippets.\n* Model identifier, max tokens \/ temperature style parameters.<\/p>\n\n<p>Headers:\n* <code>x-api-key<\/code> (your key) and <code>anthropic-version<\/code> (currently 2023-06-01 set in code).<\/p>\n\n<p>Retention (Your Server):\n* Same as OpenAI notes for conversation logging (the provider choice does not change local storage schema).<\/p>\n\n<p>Legal \/ Docs:\n* Terms: https:\/\/www.anthropic.com\/legal\/terms-of-service\n* Privacy: https:\/\/www.anthropic.com\/legal\/privacy\n* Usage Policy: https:\/\/www.anthropic.com\/legal\/aup<\/p>\n\n<p>Opt\u2011Out \/ Control:\n* Leave the Claude API key blank; those bots will fallback\/refuse if provider requires it.\n* Switch provider per bot back to OpenAI.<\/p>\n\n<h3>3. Pinecone (Optional Remote Vector Store)<\/h3>\n\n<p>Only used if you explicitly create a Context with remote type \"Pinecone\". Local context mode (default) never contacts Pinecone.<\/p>\n\n<p>Endpoint Pattern:\n* Region\/index specific HTTPS endpoints you enter (example placeholder: https:\/\/controller.pinecone.io and index query\/upsert endpoints under *.pinecone.io). The plugin validates host against an allowlist containing pinecone.io (filter extendable).<\/p>\n\n<p>Used For:\n* Upserting embeddings (during context indexing \/ syncing)\n* Querying similar vectors when answering a question in that context.<\/p>\n\n<p>Data Sent:\n* Vectors (embedding arrays) and associated metadata (post\/page IDs,  &hellip;<\/p>","raw_excerpt":"The best AI Chatbot for WordPress. Like having ChatGPT trained on your content \u2014 turn your site into a 24\/7 sales &amp; support machine.","jetpack_sharing_enabled":true,"_links":{"self":[{"href":"https:\/\/id.wordpress.org\/plugins\/wp-json\/wp\/v2\/plugin\/254489","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/id.wordpress.org\/plugins\/wp-json\/wp\/v2\/plugin"}],"about":[{"href":"https:\/\/id.wordpress.org\/plugins\/wp-json\/wp\/v2\/types\/plugin"}],"replies":[{"embeddable":true,"href":"https:\/\/id.wordpress.org\/plugins\/wp-json\/wp\/v2\/comments?post=254489"}],"author":[{"embeddable":true,"href":"https:\/\/id.wordpress.org\/plugins\/wp-json\/wporg\/v1\/users\/estebandezafra"}],"wp:attachment":[{"href":"https:\/\/id.wordpress.org\/plugins\/wp-json\/wp\/v2\/media?parent=254489"}],"wp:term":[{"taxonomy":"plugin_section","embeddable":true,"href":"https:\/\/id.wordpress.org\/plugins\/wp-json\/wp\/v2\/plugin_section?post=254489"},{"taxonomy":"plugin_tags","embeddable":true,"href":"https:\/\/id.wordpress.org\/plugins\/wp-json\/wp\/v2\/plugin_tags?post=254489"},{"taxonomy":"plugin_category","embeddable":true,"href":"https:\/\/id.wordpress.org\/plugins\/wp-json\/wp\/v2\/plugin_category?post=254489"},{"taxonomy":"plugin_contributors","embeddable":true,"href":"https:\/\/id.wordpress.org\/plugins\/wp-json\/wp\/v2\/plugin_contributors?post=254489"},{"taxonomy":"plugin_business_model","embeddable":true,"href":"https:\/\/id.wordpress.org\/plugins\/wp-json\/wp\/v2\/plugin_business_model?post=254489"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}