MxChat Core

The free MxChat plugin — the AI chatbot itself, plus everything that powers it: setup, AI models, the knowledge base, behavior, actions, display, and the REST API. Add-ons extend this core; this page documents the core itself.

Overview

MxChat turns any WordPress site into an AI chatbot that answers from your content. It runs on your server, connects to the AI provider of your choice, and answers visitor questions using a knowledge base you control — no per-message fees, no data leaving your stack except the calls you make to your chosen AI provider.

Core capabilities:

Bring-your-own AI — OpenAI, Anthropic Claude, Google Gemini, xAI Grok, DeepSeek, or any OpenAI-compatible endpoint (Ollama, LM Studio, vLLM, Azure OpenAI).
RAG knowledge base — index your pages, posts, PDFs, and custom text so the bot answers from your real content.
A guided setup wizard that walks you from "just installed" to "live on my site."
Actions — let the bot do things (capture emails, search the web, generate images, hand off to a human) not just answer.
A REST API for headless access to transcripts and knowledge.

Getting Started

When you first open MxChat in your WordPress admin, you land on the Onboarding wizard. It walks you through setup one step at a time:

Chat model — pick your AI provider and model, and paste your API key inline.
Behavior — tell the chatbot how to act (its role, tone, and the topics it should stick to). Optional — you can use the AI's defaults and refine later.
Embedding model — pick the model used to index your knowledge base. If it's the same provider as your chat model, your key is reused automatically.
Knowledge base — optionally add your first content so the bot has something to answer from. You can skip this and build it later.
Actions — optionally enable Actions (see below). Skippable.
You're set up — test the bot in your dashboard, turn it on for your site, or grab an embed shortcode.

You can reopen the setup guide any time from Settings → Tutorials.

AI Models & Providers

Choose your chat model under Settings → Chatbot → AI Models → Chat Model. From 3.2.26 the picker offers only current models, and each provider's first entry, marked Recommended, is its fast, affordable choice for customer chat. MxChat supports:

OpenAI — GPT-6.1 Sol (new in 3.2.26 and the recommended OpenAI choice), GPT-6 Astra, GPT-6 Sol and GPT-6 Luna, and the GPT-5.6 family (Sol, Terra and Luna) plus GPT-5.5. The request settings each one requires are applied automatically, and from 3.2.26 OpenAI requests are laid out so more of each prompt is served from OpenAI's prompt cache.
Anthropic Claude — Claude Haiku 5.5 (new in 3.2.26 and the recommended Claude choice), Claude Sonnet 5.5, Claude Opus 5.5 and Claude Fable 5.1, plus the previous Sonnet 5, Fable 5, Opus 5, Opus 4.8 and Haiku 4.5.
Google Gemini — Gemini 3.8 Flash (the recommended Gemini choice), Gemini 4 Argon (new in 3.2.26, Google's flagship, priced above Flash), Gemini 3.1 Pro and Gemini 3.5 Flash-Lite (the lowest-cost choice). Older Flash models (3.7, 3.6, 3.5, 3 Flash) and Gemini 2.5 are no longer offered for new picks but keep working on sites that saved them.
xAI Grok — Grok 4.20 (Non-Reasoning) (the recommended Grok choice, answers without a reasoning pause), Grok 4.7 (new in 3.2.26), Grok 4.6, 4.5, 4.3 and Grok 4.20 (Reasoning).
DeepSeek — DeepSeek Flash (new in 3.2.26 and the recommended choice; DeepSeek now answers the older V4 Flash name with it) and DeepSeek V4 Pro.
Custom (OpenAI-compatible) endpoints for local or self-hosted models.

DeepSeek users: DeepSeek retired its older API model names on July 24, 2026, which broke chat on every site still pointing at one of them. Core 3.2.15 replaced them with V4 Flash and V4 Pro, and a site still set to a retired DeepSeek model is moved to V4 Flash automatically when it updates — so the fix is simply to update to 3.2.15 or later. If DeepSeek chat is failing on an older version, that's the cause.

OpenAI GPT-5.1 Chat Latest and GPT-5.3 Chat users: OpenAI retires both models on August 10, 2026. Core 3.2.16 removes them from the picker, and a site still set to either one switches to GPT-5.6 Sol automatically, with an admin notice explaining the change. Same shape as the DeepSeek migration above — updating is the fix.

Recommended models and automatic replacement (3.2.26)

Updating to 3.2.26 never changes the model you already picked; only the Recommended label moves. The picker now holds three kinds of model apart:

Current models are offered for new picks.
Older models a provider still serves are no longer offered, but a site that saved one keeps it, and Settings still shows it as that site's value.
Models with an announced shutdown date are no longer offered from the day the date is announced. They keep working until that date, and on the date MxChat moves the saved setting to the named replacement by itself, with no update needed.

When a provider has shut a model down, MxChat switches to its replacement and shows one notice naming each old and new model, for example "Your chatbot was using Claude 3 Opus, which Anthropic has discontinued; it now uses Claude Opus 5.5." This covers the chat model, the Content screen's image model and the models add-ons store. Embedding models are never switched for you, because stored vectors only match the model that made them: when an embedding model is discontinued, MxChat shows a notice naming the model to re-embed with instead.

Saving settings keeps your models (3.2.27). Saving any settings screen no longer swaps a chat model the picker does not recognise, or an older embedding model, for a different one. Retired chat and image models still use their replacement on each request, and every model picker in the admin now reads from one model list, so the same models appear everywhere. The same release fixes o-series models, streaming through a Custom Provider, and web search queries on newer models.

Gemini Embedding 2 is new in 3.2.26 as an embedding choice (1536 dimensions). The older Gemini Embedding stays on sites that use it; Google has announced its shutdown for May 14, 2028, so it is no longer offered for new picks.

The daily model-liveness check (3.2.19)

Retirement dates only help for retirements a provider announced in advance. From 3.2.19 MxChat also checks once a day that the model you have selected still exists, and warns you on the AI Models screen and the dashboard before a retired model can silence your chatbot. Sites still set to the retired Claude Opus 4.1, or to older retired Claude models, are moved to the current Claude generation of the same tier automatically and shown a notice.

Claude prompt caching (3.2.19)

Claude API costs drop on most sites from 3.2.19 — MxChat now uses Anthropic's prompt caching automatically, with nothing to configure. This landed ahead of Anthropic's 1 September Claude Sonnet 5 price change, so the saving offsets part of that increase for sites on Sonnet.

Gemini thinking level (3.2.22)

Gemini 3-series models reason before they answer, and that hidden reasoning is billed as output tokens. From 3.2.22 MxChat asks those models for a low thinking level on every chat reply — natively and through the OpenAI-compatible provider — so a chatbot answering from a knowledge base stops paying for deliberation it does not need. Flash-Lite models do not think and are left alone. A developer can change it with the mxchat_gemini_thinking_level filter.

Gemini safety filter (3.2.25)

Google's safety filter can withhold a Gemini reply it rates as harmful. Earlier versions asked it to block anything rated medium risk or above, which withheld ordinary answers on some sites — a question about a reimbursement form on a medical-supplies store, for example — and the chatbot then stopped with "The response format was unexpected". From 3.2.25 MxChat asks Gemini to block only high-risk content in the four harm categories (harassment, hate speech, sexually explicit, dangerous content). A developer can tighten or loosen this with the mxchat_gemini_safety_settings filter (settings array, model id); returning an empty array sends no safety settings at all. The WooCommerce add-on (1.7.13) applies the same filter to its own Gemini requests.

Retirement dates in the model picker (3.2.16)

Providers retire models on published dates, and up to now that only became visible when chat broke. Since 3.2.16 the model picker shows the date on any model with an announced retirement — the entry reads "<model name> — retires <date>" — and the model disappears from the list once that date passes, so you can't newly select something that is already gone. A model you have already selected stays visible in Settings even after retirement, marked as retired, so the screen never silently shows you a different model than the one that is actually saved. From 3.2.26 a model with an announced date is no longer offered for new picks at all; a site that saved one sees it as "<model name> — retires <date>" until the date, then moves to the replacement (see above).

Add your API key for each provider you use under Settings → API Keys. MxChat detects a saved key and confirms it inline. You can switch models at any time; the model you pick is the one used for live chat answers.

The model list refreshes as providers ship new models. If a provider retires a model you had selected, MxChat automatically falls back to a current, working model on your next admin load. From 3.2.25 a chatbot still set to a retired model also uses that model's current replacement on every chat request, so it keeps answering even before anyone opens the admin.

If your provider starts rejecting the model (3.2.16). When a provider replies that the selected model isn't available on your key — "does not exist", "you do not have access", "permission denied", and the equivalent wording from Anthropic, OpenAI, xAI and Gemini — MxChat now raises a persistent admin notice naming the model and linking to Settings. Before 3.2.16 only the visitor in the chat window saw anything, so a bot could sit broken for days while the owner heard nothing. The notice is throttled to once an hour per model, so a busy broken bot doesn't flood the dashboard.

Custom & Azure OpenAI endpoints. Point MxChat at any OpenAI-compatible endpoint — Ollama, LM Studio, vLLM, llama.cpp, or Azure OpenAI — under Settings → API Keys → Custom Provider. Set the Base URL, API key, and auth scheme (Azure also takes an API version), then use the one-click Test Connection button to confirm it works; a built-in Azure OpenAI quick-start lists the exact values to paste. You can also route your knowledge-base embeddings through the same endpoint by enabling "use custom provider for embeddings" — indexing and live queries then use your endpoint's embedding model.

Local-only embedding setups (fixed in 3.2.17). If you run embeddings entirely through a custom provider and have no cloud key saved at all, ingestion now works end to end. Before 3.2.17 the import screens and the automatic post-sync checked for an OpenAI, Voyage or Gemini key before handing the job to the embedding layer — so a pure Ollama or LM Studio site got misleading "API key missing" errors on import, automatic indexing of saved posts silently did nothing, and a permanent model-mismatch banner sat on the Knowledge screen. All of those pre-flight checks are now aware of the custom provider, and the banner is gone.

Each provider's API key field now has a one-click Test key button (added in 3.2.11) that runs a quick, free validation so you can confirm a key works the moment you paste it, rather than finding out only when a chat fails.

Built-in Web Search. Turn on Settings → AI Models → Enable Web Search to let the chatbot ground its answers in live web results when a question needs current information. It uses the model provider's own search tool: OpenAI models search through OpenAI's built-in web search, and Google Gemini models now do the same through Google Search grounding (added in core 3.2.11 — previously this toggle only affected OpenAI and was a silent no-op for Gemini, so Gemini sites had to add the Perplexity add-on for web-grounded answers). Either way it uses the API key you already configured for that model, at no extra cost. This is separate from the Brave Web Search Action and the Perplexity add-on, which are other ways to add web search under Actions.

The Knowledge Base (RAG)

The knowledge base is what makes your chatbot know your stuff. Under Settings → Knowledge you can:

Auto-sync your posts and pages so published content is indexed automatically. From 3.2.17 a scheduled post is indexed at the moment it actually goes live, rather than being missed because it was published by the scheduler instead of by a person.
Upload PDFs and Word documents to extract and index their text — one at a time under Import Options, or in bulk from 3.2.22: choose WordPress Content with the content type set to Media and the importer lists the PDF, Word, text and Markdown files already in your Media Library (the types it can read text from; images, audio and video are not shown), so you can select a page of them and import them together. Switch on Media under Auto-Sync Settings → Advanced Custom Post Sync Settings and a supported file adds itself to the knowledge base on upload, updates its entry when replaced, and is removed when deleted. (Before 3.2.22, choosing Media listed nothing on every site.) From 3.2.24 Document Upload and PDF Upload each have an optional Source URL field: give the page the file lives on and answers drawn from it link to that page; leave it empty and answers name the file (and a PDF's page) in plain text, without a link. A PDF added through PDF Upload keeps its file name, its pages are listed under the file, and uploading the same PDF again replaces its pages instead of adding a second copy. PDFs uploaded before 3.2.24 are left as they are: upload once more, then delete the old pages.
Add manual entries — paste any text you want the bot to know.
Index content from a sitemap.

By default MxChat stores embeddings in your WordPress database. For larger knowledge bases you can switch to Pinecone or OpenAI Vector Store under the Knowledge page.

If you use OpenAI Vector Store, update to 3.2.18. On earlier versions every Vector Store search failed against OpenAI's API and the failure was swallowed, so the chatbot answered with no knowledge base at all — confidently, and with no error anywhere for you to notice. Nothing about your vector store or your key was wrong, and there is nothing to re-index: updating restores retrieval on the next question. The same release also fixes the Knowledge status readout, which reported "no vector stores configured" even when they were.

If you ever change your embedding model — including enabling a custom provider for embeddings — re-index your whole knowledge base (Knowledge → Process Selected) so your stored vectors and your query vectors come from the same model. A model mismatch between indexing and querying is the most common reason the bot seems to "ignore" content you've added.

Tune retrieval under Settings → Behavior: the Similarity Threshold controls how strict content matching is, and the RAG Sources / Chunks limits control how much context is sent to the AI per answer.

Re-import affected entries (3.2.24)

Earlier versions stored some imported text incorrectly: currency signs and accents dropped, HTML character codes left in, and right-to-left PDF text saved backwards. The Re-import affected entries card on the Knowledge screen, and the wp mxchat kb-reimport command, find the entries that still hold that text and rebuild them from their source. Check knowledge base lists them first and changes nothing; nothing is rebuilt until you confirm. Uploaded files are not kept on the server, so for those only text direction can be repaired.

Entries whose page was edited after import (3.2.25). On a site that keeps auto-sync off, an entry goes stale whenever its page is edited. The card now has a switch, Also rewrite entries whose page was edited after they were imported (off by default): with it on, any entry older than its post, page, product or Media Library file is rebuilt from that source as it reads today, whatever the card's date. On the command line, add --edited (for example wp mxchat kb-reimport --edited --dry-run). A site with auto-sync on does not need this.

Synced patterns, WoodMart HTML Blocks and Custom Product Tabs (3.2.25). WordPress Content import and auto-sync now include the text of synced patterns (reusable blocks) and WoodMart HTML Blocks placed on a page; before, that text was missing from the page's entry. Save or import the page again to update it. Saving a WooCommerce product also no longer removes its Custom Product Tabs for WooCommerce text from the knowledge base.

Knowledge list (3.2.25). The PDF filter shows PDF pages again, and uploaded files are listed by name instead of a View Source link that led nowhere.

WooCommerce product tags (3.2.24). Each product's knowledge-base text now carries a Tags: line beside its Categories: line. Re-save or re-import a product to pick its tags up.

Multi-Bot on the WordPress-database store (3.2.24). The Knowledge screen now says that content is shared with all bots. A bot gets a knowledge base of its own only on Pinecone (its own host or namespace) or an OpenAI Vector Store.

Page builders, ACF Blocks and Auto-Sync (3.2.26)

Page builders. The WordPress Content import and Auto-Sync now read the text Divi 4 and WPBakery keep in their shortcodes. For a page made with Breakdance, Oxygen, Bricks, Divi, WPBakery, Beaver Builder or Elementor, or any page whose saved text is under 200 characters, they also read the published page as a visitor sees it and use that text when it is clearly longer. Before 3.2.26, Breakdance, Oxygen, Bricks and Divi 5 pages were often stored as their title alone. Only published, public pages without a password are read this way. A developer can switch the page read on or off per post with the mxchat_index_use_rendered_page filter.

ACF Blocks. Text held in ACF Blocks is now indexed: each acf/ block, nested ones included, adds one "Label: value" line per field, your ACF Fields exclusions apply, and the import list's word count includes it. If you added a mxchat_before_process_post snippet to index ACF Blocks yourself, remove it, or the text is stored twice.

Rebuilding older entries. Re-import affected entries (see above) now also finds pages that were stored without their text and products missing their custom tab text. Run Check knowledge base after updating.

Auto-Sync. Each post type on Knowledge → Auto-Sync now has a Keep imported entries up to date only switch: with it on, Auto-Sync updates or removes entries already in the knowledge base but never adds new ones, so only what you imported yourself stays in sync. A save now sends one embedding request instead of two or three, and none when the text did not change. Editing a synced pattern re-syncs the pages that use it.

Leaving posts out (developers). Two filters decide whether a post goes into the knowledge base. Every indexing route applies them, in this order: the WordPress content importer, Auto-Sync (saves, publishes, scheduled publishes, a synced pattern's re-sync, WooCommerce product updates) and the Re-import tool.

mxchat_should_index_post (since 3.2.26). Arguments: $index (bool, default true), $post (WP_Post), $context (string: import, auto_sync or reimport, naming the route). Return false to leave the post out; return $index otherwise. Media Library files go through it too.
mxchat_before_process_post. Arguments: $post (WP_Post), $bot_id (string, default on Auto-Sync). Return the post, changed or not, to index it. From 3.2.26, return false to leave it out.

Only an exact false excludes, on both filters, so a callback that forgets to return a value never empties the knowledge base. The importer reports an excluded post as skipped ("excluded by a filter"), and when Auto-Sync meets a post a filter now excludes it removes the entry that post already had, so the knowledge base follows your rule.

// Keep a category and anything marked private out of the knowledge base, on every route.
add_filter( 'mxchat_should_index_post', function ( $index, $post, $context ) {
    if ( has_category( 'internal', $post ) || get_post_meta( $post->ID, 'kb_private', true ) ) {
        return false;
    }
    return $index;
}, 10, 3 );

// Or skip a post from the filter you may already use to change post data (3.2.26+).
add_filter( 'mxchat_before_process_post', function ( $post, $bot_id ) {
    if ( 'draft-notes' === $post->post_type ) {
        return false;
    }
    return $post;
}, 10, 2 );

Sync fixes (3.2.27)

Knowledge base sync now follows synced patterns, posts that are unpublished, trashed or restored, and PDFs placed in ACF Blocks. In the WordPress Content import list, pages built with Breakdance, Oxygen, Bricks or Divi 5 now show their published text, the same text the import stores.

One embedding call per reply. Each visitor message used to be turned into an embedding twice, once to match Actions and once to search the knowledge base. From 3.2.27 the search reuses the first one when the text is the same, and with an OpenAI Vector Store serving the turn (which searches by text) no query embedding is made unless the visitor uploaded a PDF or Word file in the chat.

Pinecone document index with full-text search (3.2.22)

Pinecone offers two kinds of index and MxChat supports both. A vector index — the default, and what every existing setup uses — stores your embeddings and answers similarity queries. A document index with full-text search additionally keeps a keyword index of the stored text, so a question that contains a part number, SKU or code is matched on the literal string as well as on meaning. Choose the type under Knowledge → Pinecone → Index type; the classic vector path is untouched, so nothing changes for a site that leaves the default.

A document index is a separate index at Pinecone. Create it with Create index for me — a new index named after your Index Name field, sized for your embedding model, in the cloud, region and language you pick, with the host filled in for you — or enter an existing document index's host and click Check index, which refuses a classic index, a foreign schema or a dimension mismatch with the exact reason. The type only saves once the host has been checked. Migrate then copies your existing vector index into the document index — text, vectors and metadata, same ids — with no re-embedding; it runs in resumable batches and is safe to run again. Delete old index appears once the copy count matches and asks you to type the index name. Your embedding provider and model do not change.

Changed in 3.2.23. Migrate now runs while the chatbot keeps answering from the vector index: nothing switches until you click Save Pinecone Settings, and saving is held back while a copy into that index is still running unless you tick Switch anyway. An interrupted copy continues where it stopped, dropped connections are retried, the real reason is shown if it does stop, and the progress line reports records remaining, records in other namespaces, and the document index's own count when done. When the source holds several namespaces (one per bot, or one left from an earlier setup), a Copy every namespace switch beside Migrate copies each into the same-named namespace of the document index. Delete old index now checks every namespace and refuses, naming the namespace, while any namespace with records lacks a finished copy whose count matches. Switching the index type to Document index suggests a free name ending in -docs, so Create index for me no longer fails with HTTP 409 when the Index Name field still holds the vector index's name, and neither Create nor Check moves the chatbot before you save. Part numbers typed without their spaces, hyphens or dots — R165321320 for R1653 213 20 — now find the right record in any letter case, with an exact match ranked above a family match; records copied on 3.2.22 need Migrate run again to pick up the matching tokens.

With the index type on Document index, switch on Hybrid keyword boost under Chunking & Retrieval: a question carrying a code-like token gets a second lookup restricted to records containing that code (with a full-text search as the fallback), fused with the semantic results, and the Testing and Transcripts panels label each match Vector, Keyword or Both. The Multi-Bot add-on does not yet offer a per-bot index type.

Also fixed in 3.2.22: with Pinecone, the Knowledge Entries heading now shows the vector count Pinecone reports instead of "(50)" for any index over 500 vectors, paging uses Previous and Next and continues where the last page ended, and a bot with its own empty namespace shows 0.

Role restrictions: any role, and pages too (3.2.22)

Knowledge → Role Restrictions maps a post tag to the role a visitor must hold for the chatbot to retrieve that content. From 3.2.22 the role list is built from the roles your site actually has, so alongside Public, Logged In Users and the built-in Subscribers & Above through Administrators Only you can pick a custom role from User Role Editor, Members or a membership plugin; access is granted to that role or to any role whose capabilities cover it, which keeps the built-in "& Above" ladder exactly as it was. Because tags only exist on posts, every post and page also gets a Knowledge base access dropdown in the MxChat box on its edit screen (default Use tag rules), so a page or an untagged post is no longer stuck as public; where both apply, the stricter of the tag rule and the page setting wins. The same dropdown sits on each entry in the Knowledge list for content that came from a PDF, a URL or an upload. This release also closes a leak: on earlier versions a restriction set to a custom role was readable by every signed-in visitor.

Hybrid keyword boost (3.2.16)

Vector search matches on meaning, which is exactly wrong for a question built around an exact token — a product code, a SKU, an error code, a person's name. Semantically, ERR-4102 looks a lot like ERR-4103.

Knowledge → Chunking & Retrieval → Hybrid keyword boost (labelled WordPress database and Pinecone document indexes from 3.2.22) combines keyword matching with the existing similarity search so those questions find the right document. It is off by default. (In 3.2.17 this setting moved here from Settings → Behavior, alongside the chunking controls it belongs with; your saved choice carried over, and the old location links to the new one.) Two things to know before enabling it:

It affects the WordPress-database knowledge base and, from 3.2.22, a Pinecone document index (see above). On a classic Pinecone vector index or an OpenAI Vector Store the toggle does nothing.
It changes which sources are retrieved on an existing install. Ask your bot a handful of your real questions after turning it on rather than assuming it is a free win.

While it is on, the Transcripts and (from 3.2.22) Testing panels label how each match was found — Vector, Keyword, or Both — so you can see whether the keyword half is actually rescuing anything before you leave it on.

Unpublished content and the knowledge base (3.2.16)

Unpublishing, trashing, or making a post private now always removes it from the knowledge base — including status changes made by scheduled-expiry plugins that change post status directly and never touch the edit screen. Previously those could leave a chatbot answering from content that is no longer public on the site.

To clear entries left behind by earlier unpublishes, run the new WP-CLI command once:

wp mxchat prune-unpublished --dry-run
wp mxchat prune-unpublished

The --dry-run pass reports what it would remove without deleting anything.

Behavior & System Instructions

Under Settings → Behavior, the AI Instructions field is the chatbot's "job description" — the instructions it follows on every reply (its role, tone, and guardrails). This is the highest-leverage setting for making the bot sound like your brand.

Since 3.2.15 you can also decide where your knowledge-base content is inserted into those instructions, using the context placeholder — put it wherever the retrieved content should sit, so you can write rules that come after the source material. Leave the placeholder out and MxChat behaves exactly as before.

Since 3.2.16, when the knowledge base has nothing relevant to a question, the chatbot is told that plainly instead of being handed a "no information found" note formatted like knowledge base content. Previously the model could treat that note as source material and answer around it; now it simply says it doesn't know. If your bot used to produce oddly confident non-answers on out-of-scope questions, this is the change.

The chatbot knows what day it is (3.2.22). Every request now carries a line of the form Current date and time: Monday, September 21, 2026, 9:15 AM (Europe/Amsterdam). Anything dated before today is in the past., built from the timezone under Settings → General, so the model can tell a past event from an upcoming one without a rule from you. The line travels with the per-message context rather than inside the instructions, so it does not affect prompt caching. If you want the date at a particular spot in your own rules, {current_date} and {current_datetime} are replaced in the AI Instructions field — with the caveat the field's hint spells out: a placeholder changes the instructions once a day or once a minute and reduces how much of the prompt your provider can cache, so most sites do not need one. Developers can reword or suppress the automatic line with the mxchat_current_datetime_line filter.

Note that Actions dispatch outside the reply text, so an instruction like "never ask visitors for their email" will not reliably stop an enabled Action from firing. To stop an Action, disable it under Settings → Actions — see the FAQ entry on the mailing-list prompt.

Other behavior controls:

Contextual Awareness — let the bot reference the current page's content for more relevant answers. From 3.2.22 the page text is read from an inert copy gathered once per page: on earlier versions the copy was made live on every message, and on a page with a video or audio player it could start an invisible copy playing (with sound, if the visible player had been unmuted) and download the media again each time.
Citation Links — allow the bot to include source links from your knowledge base.
Strip Unapproved Links (3.2.20+) — remove links the AI invents, even when Citation Links is off. MxChat has always checked every URL in a reply against an approved list and dropped the rest, but until 3.2.20 that enforcement rode on Citation Links, so switching citations off also switched the guard off and let fabricated URLs through. It is now its own control. Approved citations, real pages on this site, and links returned by your integrations — WooCommerce product URLs, for example — are always kept. On by default for new installs; existing sites keep their current behaviour until you turn it on.
Satisfaction Rating — show a 👍 / 👎 prompt after a conversation so you can measure answer quality.
Message limits — cap how many messages visitors can send per hour, day, week, or month. Set a limit per user role (including logged-out visitors), or a single total cap across everyone, each with a live usage readout ("10 of 200 used · 190 left · resets in ~6 hours") and a one-click reset. Limits accept any whole number, not just the presets; the default is Unlimited, so existing sites are unchanged until you set one. From 3.2.22 a visitor's allowance is counted against the address the request really came from — Cloudflare is recognised automatically, and other proxies can supply the address with the mxchat_client_ip filter — and daily, weekly and monthly limits hold for the full period. On 3.2.21 and earlier the per-visitor key came from a client-supplied header and every counter was cleared hourly, so a daily cap effectively reset each hour; if your limits never seemed to bite, that is why. A signed-in user who holds several roles is limited by the first role WordPress stored for them; from 3.2.27 the mxchat_filter_roles_rate_limit filter (roles array, WP_User, bot id) can reorder them, and the first role it returns is the one whose limit applies. Also from 3.2.27, a visitor's first open of the chat no longer hits a Too Many Requests error.
Reply length (3.2.26) — there is no max-tokens setting; ask for shorter answers in AI Instructions (Behavior). MxChat no longer caps Claude replies at 1,000 tokens: they get 4,096 output tokens, or 8,192 for Claude Opus 5 and later and the Fable models, which think before answering (filter: mxchat_claude_max_tokens). A reply that reaches the provider's limit, with or without AI Tools, now ends with (Reply shortened: it reached the length limit.) instead of stopping mid-sentence.
Markdown in the chat window — replies (and the visitor's own messages) are formatted for bold, italic, strikethrough, headings, code and tables. From 3.2.22 words joined with underscores — file names, SKUs, snake_case identifiers, _PLACEHOLDER_ tokens — keep their underscores instead of turning italic, a backslash writes an underscore, asterisk, tilde or backtick literally, and anything inside backticks is shown exactly as written. There is no switch to turn formatting off.

Actions

Actions let your chatbot do things, not just answer — capture an email, search the web, generate an image, answer from a PDF, or hand off to a human. The free core includes:

Loops Email Capture — collect a visitor's email into your Loops mailing list.
Brave Web Search and Brave Image Search — search the web and images from chat.
Generate Image — create an image in the chat with OpenAI, Google or, from 3.2.27, xAI. From 3.2.26 OpenAI uses GPT Image 2.5 Flare (GPT Image 1 retires on October 23, 2026) and Google uses Nano Banana 2.1, which fixes Gemini image generation that had stopped working after Google shut down Imagen. Generate Image (xAI) (new in 3.2.27) uses Grok Imagine 2.0 and needs an xAI API key under Settings → API Keys; like the other two it is both an Action and an AI Tool, and it saves the image the same way the Content screen does. Developers can choose other models with the mxchat_openai_image_model, mxchat_gemini_image_model and (3.2.27) mxchat_xai_image_model filters. The Content screen's blog-post images now use the Image Model you choose there. From 3.2.27 a retired image model is swapped for its replacement on every request.
Chat with PDF — answer questions about an uploaded PDF.
Live Agent Handoff — transfer the conversation to a human on Slack or Telegram, and hand it back to the bot when done.

There are two ways the bot decides when to run one of these, both under Settings → Actions:

AI Tools (function calling) — the model decides on its own when to use an enabled tool, straight from the visitor's natural-language message, with no trigger phrases to maintain. Enable individual tools under Actions → AI Tools. AI Tools is off by default, and every tool is turned on deliberately — sensitive ones (anything that spends money, exposes customer data, or hands off to a human) stay off until you switch them on. Each tool can take an optional hint that tells the model when to reach for it; from 3.2.19 the two live-agent handoff tools arrive with that hint prefilled, so automatic escalation works as soon as you enable them, and you can reword or clear it. AI Tools needs a tool-capable chat model; if your current model can't call tools, the page says so and the tools simply won't fire until you pick one that can. From 3.2.23 the model is told how many tool calls it may make for the message and how many remain after each round, so it no longer asks for several lookups when only one is allowed and answers from partial results. From 3.2.22 a question that needs no tool is generated once instead of twice, so ordinary turns arrive sooner and use roughly half the output tokens; those replies arrive as one message, and the mxchat_fc_keep_first_hop_text filter restores the old two-part behaviour. The same release fixes the loop sending an invalid request once the per-turn tool budget was spent — every tool call now gets a result, the model answers from what it has, and a provider failure after a tool has run is logged even with developer mode off.

Seeing which tools ran (3.2.19)

Until 3.2.19 the Message Context panel in Transcripts reported only how Trigger Phrases scored, so a site running purely on AI Tools saw "No action analysis available for this message" while its tools were firing correctly. From 3.2.19 each message also records which AI Tools ran, how long each took, and the error text if one failed. Tools that handle money, customer records or a live-agent handoff record only the fact that they ran, never their arguments — the fired-fact is the half you need for debugging, and the arguments are the half that carries personal data. The record is written as the answer is generated and is never backfilled, so it does not appear on conversations from before you enabled the tool.
- Trigger Phrases — the original approach: you list the phrases that should fire an Action, and a matching message runs it. Found under Actions → Trigger Phrases (this is the renamed home of the phrase-matching system).

You can run both at once — when a visitor's message matches a trigger phrase that fires first, and AI Tools cover everything else. Many add-ons (WooCommerce, Forms, Perplexity, and more) register additional Actions, and those become available as AI Tools too.

Brave Web & Image Search

The web and image search Actions run on the Brave Search API. Brave operates its own independent web index rather than reselling another engine's results, which is why these are separate from the Enable Web Search toggle under AI Models (that one uses your model provider's built-in search) and from the Perplexity add-on. Both Brave Actions ship with the free core plugin, and both need a Brave Search API key before they will return anything.

Step 1 — Get a Brave Search API key

Create an account at api-dashboard.search.brave.com.
Subscribe to a Search API plan. Brave offers a free tier that is enough to try the feature out.
Copy your subscription token.

Step 2 — Add the key in WordPress

Paste it into Settings → API Keys → Brave API Key and save. The Settings → Brave Search tab holds the search options described below, but the key itself lives on the API Keys tab alongside your model keys.

Step 3 — Turn the Actions on

Under Settings → Actions → AI Tools, in the Search Features group, enable:

Brave Web Search — searches the web and answers from the results.
Brave Image Search — returns images inline in the conversation.

AI Tools is off by default and every tool is enabled deliberately, so nothing searches until you switch it on. AI Tools also needs a tool-capable chat model. If you would rather match on wording than let the model decide, both Actions can be fired from Actions → Trigger Phrases instead.

If either Action is enabled while no Brave key is saved, the Actions page shows a warning — without the key the tool is registered but silently returns nothing.

Search options

Under Settings → Brave Search:

Number of Images to Return — how many images an image search shows. Defaults to 4.
Safe Search — filters adult content in image results: Strict (the default) or Off.
Number of News Articles — how many news results to include.
Country — biases results toward a country.
Language — biases results toward a language.

Web search returns five results per query.

How a search actually runs

The query is rewritten first. MxChat does not send the visitor's raw message to Brave. Your selected chat model interprets it into a focused search query, so "do you know anything about what happened with that new pricing thing" becomes something Brave can actually match. If the model cannot make a query out of the message, the bot asks the visitor to rephrase instead of searching blindly. Before 3.2.25 this step returned nothing on Gemini 3 Flash and Pro models (the model spent its small output allowance on thinking), so the visitor's whole sentence was sent to Brave; from 3.2.25 those models produce key words like every other provider.
Brave is called server-side — api.search.brave.com/res/v1/web/search for web and /res/v1/images/search for images, authenticated with your subscription token in the X-Subscription-Token header. The key never reaches the browser.
Results are cached. Each distinct query is stored in a WordPress transient, so repeated questions do not spend quota twice.
The model answers from the results, with links, rather than pasting a raw result list.

Troubleshooting Brave search

The bot never searches. Check that the Action is enabled under Actions → AI Tools, that a Brave API key is saved under API Keys, and that your chat model supports tool calling.
Results look stale after changing settings. Searches are cached per query. Try a differently worded question, or wait for the cached copy to expire.
Image results are missing or filtered. Safe Search defaults to Strict. Set it to Off under Settings → Brave Search if the topic legitimately needs it.
Nothing returned for an obscure query. The rewritten query is what reaches Brave; check the Transcripts → Message Context panel to see which tools ran.

Slack Live Agent Handoff

The Live Agent Handoff Action lets a visitor escalate from the AI to a real person. Your team replies inside Slack and every reply is relayed straight back to the visitor's chat window — there's no separate agent dashboard to watch.

How it works: a visitor's message matches a trigger phrase, so MxChat opens the conversation in Slack, invites your agent User IDs, and posts a New Live Agent Request with the session ID and the last few messages of context. Agents reply there, and MxChat relays each reply back to the visitor through /wp-json/mxchat/v1/slack-messages. The session stays in live-agent mode until it's handed back to the AI.

Two channel modes (the shared-channel option was added in 3.2.15). By default MxChat creates a new public channel per conversation, named after the chat session. Since 3.2.15 you can instead point every handoff at one shared channel your team already monitors, with one thread per visitor — useful if per-visitor channels were cluttering your workspace. Set the shared channel in the Slack settings; leave the field blank and behavior is unchanged.

Where to reply (changed in 3.2.21). In a shared channel every conversation lives in a thread rooted at its handoff message, and agents must reply inside that thread — a channel-level reply there reaches nobody. In a per-conversation channel, agents can reply in the channel or in a thread under the handoff message; on 3.2.20 and earlier a threaded reply in a per-conversation channel was silently discarded, and from 3.2.21 it reaches the visitor. Messages Slack tags with a subtype — an edited reply, a file, snippet or image upload — are never relayed; from 3.2.21 the agent sees a short note in the channel explaining how to get the message through, and the drop is logged when Debug Mode is on. If your shared channel is private, Slack sends replies as message.groups events, which the app manifest does not subscribe to: add the groups:history scope, subscribe to message.groups alongside message.channels, and reinstall the app. From 3.2.21 the Slack settings screen warns you when the shared channel it is pointed at is private.

Also added in 3.2.15: an optional setting to archive a conversation's Slack channel when an agent ends the chat, so per-visitor channels stop piling up. It is off by default.

Step 1 — Download the Slack app manifest

Download the ready-made app manifest. It pre-fills the bot user, permission scopes, and request URLs so you don't have to configure them by hand:

Download website-to-slack.txt

Step 2 — Create the Slack app from the manifest

Go to api.slack.com/apps and click Create New App → From a manifest.
Choose your workspace, open the YAML tab, and paste the contents of website-to-slack.txt.
Replace both https://example.com placeholders with your own domain, so the URLs read https://YOUR-SITE.com/wp-json/mxchat/v1/slack-messages and https://YOUR-SITE.com/wp-json/mxchat/v1/slack-interaction.
Click Create, then Install to Workspace and approve the permissions.

Step 3 — Copy your Bot Token

Open OAuth & Permissions and copy the Bot User OAuth Token — it starts with xoxb-. While you're in the app settings, also grab the Signing Secret from Basic Information → App Credentials.

Step 4 — Connect Slack in WordPress

In your WordPress admin, go to Settings → Integrations → Slack and fill in:

Live Agent Status — turn it On.
Slack Bot OAuth Token — the xoxb-… token from Step 3.
Slack Agent User IDs — the member ID of each agent who should be pulled into chats, one per line (e.g. U1234567890). In Slack, open a profile, then More → Copy member ID.
Slack Secret Key — your app's Signing Secret; it verifies that incoming requests really came from Slack.
Slack Webhook URL — your Incoming Webhook URL, used for agent notifications.
Notification Message — what the visitor sees once an agent has been pinged.
Away Message — shown when Live Agent is off, so the bot keeps helping on its own.

Then click Test Slack Connection to confirm the token authenticates and can create channels.

Step 5 — Turn on the handoff Action

Go to Settings → Actions, enable Slack Live Agent, and add the trigger phrases that should escalate a chat — for example "talk to a human", "speak to an agent", or "live support". Save, and the handoff is live.

Alternatively, enable Hand Off to Live Agent (Slack) under Actions → AI Tools and let the model decide when to escalate, with no phrase list to maintain. From 3.2.19 that tool ships with its usage note already filled in — use this when the visitor has asked something you cannot answer from the knowledge base, or has asked twice about the same unresolved problem — so switching it on is enough on its own.

What the agent sees

The New Live Agent Request posted into Slack carries the session ID, the WordPress user ID, the visitor's name and email if either was captured, then the last five messages of the conversation as a quoted Recent Conversation block, and finally the message that triggered the handoff. It is the last five turns rather than the whole transcript — open MxChat → Transcripts and search the session ID printed in the Slack message for the full conversation. From 3.2.21 the count is filterable with mxchat_handoff_history_count (one filter for Slack, Telegram and the webhook destination; Telegram asks for 10).

Required OAuth scopes

The manifest already grants everything the integration needs. If you ever build the app by hand instead, add these Bot Token Scopes under OAuth & Permissions → Scopes, then reinstall the app:

channels:manage · channels:read · chat:write · chat:write.public · groups:write · users:read

The manifest grants a slightly wider set — it also adds commands, im:write, incoming-webhook, app_mentions:read, channels:history, and groups:read — so every handoff feature works out of the box. When in doubt, use the manifest.

Troubleshooting Slack handoff

Nothing happens when a visitor asks for an agent — confirm Live Agent Status is On, the Bot OAuth Token is saved, and the Slack Live Agent Action has trigger phrases. Run Test Slack Connection.
Agents aren't added to the channel — make sure the Agent User IDs are member IDs (they start with U), one per line, and that the app has the channels:manage and channels:read scopes.
Agent replies don't reach the visitor — in Slack, check that the Event Subscriptions request URL points to https://YOUR-SITE.com/wp-json/mxchat/v1/slack-messages and shows as Verified. If that is right and replies still vanish, see Where to reply above: a thread reply in a per-conversation channel (fixed in 3.2.21), an edited or file-based reply, or a private shared channel without the message.groups subscription.

Telegram handoff

The Telegram destination has its own guide — MxChat Telegram Integration — covering the bot, the supergroup with Topics, and the webhook secret. What changed in 3.2.21: the handoff now shares the last 10 messages (up from five), labelled with the count and split into several Telegram messages when the history is long, where before a conversation over Telegram's message limit could drop the whole handoff card without a trace; ending a conversation — with #close in the topic or by closing the topic in Telegram — now retires that topic, so the visitor's next request opens a fresh one instead of posting into a closed topic agents cannot reply in; any message Telegram refuses to deliver is shown on the Telegram settings screen with Telegram's own explanation; and when your site rejects Telegram's incoming requests — the usual cause is Cloudflare or a reverse proxy in front of a webhook registered without a secret — the Telegram settings screen and a dashboard notice say how many were rejected, when, why, and how to fix it. The Open Telegram settings button on the webhook-security notice now opens the Telegram tab rather than the onboarding screen.

Displaying the Chatbot

Turn the chatbot on for your whole site under Settings → Display → Auto-Display Chatbot, or place it manually with a shortcode:

[mxchat_chatbot floating="yes"] — a floating widget in the bottom corner.
[mxchat_chatbot floating="no"] — an inline embedded chat anywhere on a page.

The Display tab also controls chat persistence, link behavior, the chat window's text and labels, and appearance. To deploy the same chatbot on a non-WordPress site, see the MxChat Anywhere add-on.

Visitors can save a conversation with the Download Transcript item in the chat window menu. Show or hide it under Settings → Display → Visibility → Show Download Transcript Button (on by default). You can also add a Start new chat item to the same menu — it clears the current conversation and begins a fresh session (after a quick confirm) — by turning on Settings → Display → Show Start-New-Chat Button (off by default; rename the item with the adjacent label field). With Chat Persistence on, 3.2.21 fixes what a returning visitor sees: Start new chat now leaves only your greeting, and a restored conversation keeps the greeting above it — on earlier versions a returning visitor could be left with one old reply and no greeting. While the bot is streaming an answer, the send button becomes a Stop control — click it to end the stream and keep the partial reply.

Pre-chat lead capture. Settings → Chatbot → Lead Capture → Require Email to Chat puts a name-and-email form in front of the first message. Since 3.2.20 the form can carry a Consent Checkbox with your own label and Privacy Policy link, optionally required and enforced on the server, and each capture records whether it was ticked, when, and the exact label shown — visible in the Leads tab, the CSV export and privacy exports. From 3.2.22 two more toggles on the same screen: Also Show for Logged-In Users shows the form to signed-in visitors too, pre-filled from their profile (by default they skip it, which on a membership or LMS site meant no lead and no consent was ever captured), and Require Email Address off makes the email field optional, so you can collect a name and consent alone. Leads captured either way appear in the Leads tab, the CSV export and the GET /leads API endpoint.

The floating launcher button is fully keyboard-operable: Tab to focus it, Enter or Space to open the chat, and Esc to close it — focus returns to the launcher on close (WCAG 2.1 Level A).

Loading chatbot files only where the chatbot appears (added in 3.2.15). By default MxChat loads its CSS and JavaScript on every page. Core 3.2.15 adds an optional performance setting that loads them only on pages where the chatbot actually appears — useful when a performance audit flags MxChat under "reduce unused CSS and JavaScript". It is off by default, so existing sites are unchanged until you turn it on. If you use the MxChat Triggers add-on, enabling this is safe: a trigger's job is to open the chat widget, so a trigger on a page with no widget was already non-functional.

Embedded and floating on the same page (3.2.25). On a page that shows a bot inline with [mxchat_chatbot floating="no"], MxChat no longer also adds the floating bubble of the same bot. On earlier versions that second bubble opened but could not send messages.

Chat styles stay inside the chat (3.2.25). MxChat's chat styles are scoped to the chat window, so they no longer change the rest of your site — code blocks on your own pages turning black was the common symptom.

Links in replies (3.2.25). A www address at the end of a paragraph and a web address written in bold now become correct links in the chat widget.

Page caches and Delay JavaScript (3.2.25, updated 3.2.26)

Two controls under Settings → Optimization matter on cached sites:

Page Cache Compatibility → Clear page cache when MxChat settings change — on by default. When on, MxChat clears the page caches it can reach (WP Rocket, LiteSpeed Cache, W3 Total Cache, WP Super Cache, FlyingPress) after its settings are saved, at most once every 30 seconds, so visitors do not keep seeing the old chatbot. Turn it off on a large site where every save would throw away the whole cache; you then clear the page cache yourself after changing the chatbot. Developers can narrow which settings trigger the clear with the mxchat_cache_purge_watched_options filter. The option is stored as mxchat_cache_purge_enabled.
Script Loading — from 3.2.26 the chat launcher is part of the page itself, so it appears straight away on cached pages and when a cache or performance plugin's Delay JavaScript holds back chat-script.js. The chat opens once that script has run; a tap before then does nothing, so if you want the very first tap to open the chat, leave chat-script.js out of the delay. With the Complianz Integration switch on (Settings → Chatbot → Lead Capture → Privacy & Compliance), the launcher stays hidden until the visitor consents, and from 3.2.26 that also holds with deferred, delayed or on-interaction script loading.
MxChat's own Script Loading Strategy (3.2.27) — with a Delay or On User Interaction strategy under Optimization & Diagnostics → Script Loading, a click (or Enter or Space) on the launcher before the chat script has loaded is remembered, starts the load, and opens the chat once it is ready, so one click is enough. A cache plugin's Delay JavaScript is a separate thing: there, a tap before chat-script.js runs still does nothing.

With the Complianz Integration switch on, the chat never opens without the visitor's marketing consent. From 3.2.27:

Chats placed in a page with [mxchat_chatbot floating="no"] wait for consent like the floating chat. Until then the page shows "The chat will appear here once you accept the cookies it needs." in place of the chat, and nothing of the chat runs.
A Complianz script that loads late is waited for, instead of the chat deciding before Complianz has answered.
Withdrawing consent closes the chat, and an embedded chat goes back to its placeholder.

MxChat Pro on each site (3.2.25)

A licence holder can turn MxChat Pro on or off for each site activated on their key from the Licenses page of their mxchat.ai account — useful for an agency that resells to clients. The site learns of the change by checking with mxchat.ai:

A site that is on checks once a day, at a random time per site.
Only an explicit "off" counts. A timeout, a server error, an mxchat.ai outage or an unreadable reply never switches a site off, and neither does a site move: a valid key on a domain mxchat.ai does not list stays on.
Three-day grace period. After the first "off" the site's admin sees a notice saying when Pro will pause. Pro is paused only if the domain is still off three days later; switching it back on before then changes nothing.
What pausing does. Only Pro features stop. Settings, the knowledge base, the saved licence key and email, and the free features are untouched, so switching back on restores Pro in place.
Switching back on is quick. A site that is off or in its grace period re-checks when an admin opens any MxChat screen (at most once a minute) and hourly in the background, so Pro returns within the hour. The Check license now button on MxChat → Pro & Extensions checks immediately.

The licence status on Pro & Extensions shows the current state — Active, Switching off, Switched off or Revoked — with the date Pro was or will be paused. The site cannot switch itself back on; only the licence holder can, in their mxchat.ai account. Each check sends the licence key, account email, site host name and plugin version to mxchat.ai.

WordPress ships its own data-subject tooling at Tools → Export Personal Data and Tools → Erase Personal Data. Since core 3.2.16, MxChat registers with it — so when you run a request against your site, chat conversations are included automatically. There is nothing to configure and no separate MxChat screen to remember.

This matters because chat is a free-text field. Visitors paste order numbers, addresses and account details into it, and MxChat also records the email and name captured during a conversation, the page the chat started on, and — for anonymous visitors — the IP address used as their identifier. Before 3.2.16 a site owner could run WordPress's own tool, get a clean report, and send it while every chat transcript for that person sat untouched.

How a person is matched. WordPress hands the tools an email address. MxChat matches conversations three ways:

the email captured on the conversation matches the address, and
when that address belongs to a registered WordPress account, the conversation's user ID or username matches that account.

The second rule is what catches logged-in conversations where the visitor never typed their email into the chat. The practical limit is worth knowing: a purely anonymous visitor who never gave an email and has no account is identified only by IP, and WordPress's privacy tools are keyed by email address — so there is no handle to match them on. That is a property of the WordPress tooling, not of MxChat.

What an export contains. One item per conversation, grouped under Chat Conversations (MxChat): the session ID, when it started and ended, the page it started on, any email/name/identifier on record, and then every message in order, each labelled with its timestamp and whether it came from the visitor, the AI, or a live agent. Exports are paginated (10 conversations per request), so a heavily-used site completes without timing out.

What an erasure removes. For each matching conversation: the transcript rows, any stored translations of them, the saved conversation history and captured name/email held against that session, and the temporary records from Chat with PDF / Word document uploads — including deleting the uploaded file itself from disk. Erasures run 25 conversations per batch.

Keeping a conversation on legal hold. If your site has to retain a specific conversation despite an erasure request, filter it:

add_filter('mxchat_privacy_erase_session', function($erase, $session_id, $email) {
    // return false to retain this conversation
    return $erase;
}, 10, 3);

A retained conversation is reported back in the erasure result, never silently kept — WordPress's report will say it was retained by a site-specific policy, which is what you need for your own records.

Suggested privacy-policy text. MxChat also adds a suggested section to Settings → Privacy → Privacy Policy Guide, describing what the chat stores, that it stays in your own database rather than a third-party chat service, how long it is kept, and that it is covered by export and erasure requests. Copy it into your policy and edit to taste.

This is separate from transcript retention (MxChat → Transcripts → Notifications, the Transcript Management card), which deletes conversations in bulk after a set period. Retention is a time-based sweep; this is a named individual's access or erasure request. You want both.

Delete Leads After (3.2.27)

Transcript retention removes conversations but keeps the lead. To remove leads too, use Delete Leads After on the same Transcript Management card (MxChat → Transcripts → Notifications). It deletes a lead (the visitor's name, email and consent record) and its conversations once the visitor has not chatted for the period you choose:

Never (Keep All) — the default; nothing changes until you pick another value.
Same as transcripts — follows the transcript retention, including Custom Retention (Days) and the mxchat_transcript_retention_days filter.
After 30, 90, 180 or 365 days.

A lead that is still chatting is never removed: the period counts from the visitor's latest message or session activity. The sweep runs with MxChat's daily maintenance, up to 500 leads per run, and each lead is removed the same way a WordPress erasure request removes it. Bans and Forms submissions are kept. While a period is set, the Leads screen says how many days leads are kept. Developers can change the period with the mxchat_lead_retention_days filter (days, saved choice); return 0 to keep leads.

Also from 3.2.27, deleting a lead or a chat removes its ratings and link clicks with it.

Transcripts: tokens and staff access (3.2.26)

Tokens per reply. MxChat → Transcripts now shows the model each reply used and its tokens, for example 1,240 in (1,024 cached), 310 out, with a total for the conversation, so you can see what a chat cost and how much your provider served from its prompt cache.

Staff access. A user without administrator rights can read, search and export transcripts and leads if their role has the mxchat_view_transcripts capability; add it with a role editor plugin (administrators can always open Transcripts). Such a user sees only the Transcripts menu, and deleting conversations or changing transcript settings still needs an administrator. A developer can point the screen at a capability staff already have with the mxchat_transcripts_capability filter.

Also in 3.2.26. Ratings and feedback are deleted and exported together with their conversation, search replies are saved to the transcript once instead of twice, and guest IP addresses show in transcript details.

Tokens spent finding the answer (3.2.27). Below a reply's own tokens, Transcripts now add a line such as plus retrieval/search 2,134 in, 85 out: the tokens your provider reported for finding that answer, which covers turning the question into an embedding for the knowledge base search, an OpenAI Vector Store search, rewriting the question into a search query, and product searches an add-on such as WooCommerce ran. They are kept apart from the reply's numbers. Replies saved before 3.2.27 show no such line.

Replies as the visitor saw them (3.2.27). Transcripts, a reloaded chat and the transcript email now show each reply as one message the way the visitor saw it, with its product cards, forms and images, plus the pre-chat name and email, and without stray code or empty messages.

Session storage and automatic cleanup (3.2.18)

Every conversation carries a little working state — which bot it belongs to, a captured name or email, live-agent status. Until 3.2.18 each of those was written to the WordPress options table as its own row and never cleaned up, so a busy site accumulated thousands of them and some were loaded on every single page view. From 3.2.18 that state lives in its own table, {prefix}mxchat_sessions, and existing rows are moved across automatically in background batches after you update — there is nothing to run and no setting to change.

Sessions untouched for 30 days are now deleted on their own. Two things trigger the sweep, deliberately: a daily maintenance event, and a capped opportunistic pass on the write path that runs at most once an hour and removes a bounded number of rows. That second route is what makes the cleanup survive a site where WP-Cron never runs — a common state on hosts that disable it without substituting a real system cron.

To keep sessions longer or shorter, filter the window:

add_filter('mxchat_session_retention_days', function($days) {
    return 90;
});

A value of zero or less is ignored rather than obeyed, since it would sweep conversations that are still live. This is about the chatbot's own per-session working state — your transcripts are governed separately under Settings → Transcripts.

REST API

MxChat exposes a bearer-token-authenticated REST API for headless access. Generate and manage your token under Settings → API Access. Endpoints: GET /health, GET and DELETE /transcripts, POST /knowledge to push content, and from 3.2.22 DELETE /knowledge — remove entries by source_url, singly or up to 50 at a time, so a document deleted in SharePoint, Drive or your CMS can be taken out of the chatbot by your sync job (a URL that is not stored is reported as nothing removed, not an error) — and GET /leads, which reads lead-capture submissions with paging, a since filter, search and a status filter, so leads can be synced into a CRM or spreadsheet automatically. The API Access screen documents every endpoint in one place with ready-to-run examples for both new calls. The chat widget itself also fetches a fresh per-request security token from /wp-json/mxchat/v1/nonce, which keeps the first message reliable even behind full-page caches.

Transcripts endpoint (3.2.27). GET /transcripts takes a new knowledge_used filter (yes, no or any): knowledge_used=no returns every answer given without knowledge base content, both those with no lookup recorded and those whose lookup used no source. has_rag_context=no misses the second kind. Each row now also reports sources_used (a number, or null when no lookup was recorded) and knowledge_used (true or false). Filtering by role=assistant now returns the chatbot's replies; they come back with role bot (role=bot selects the same rows), and live-agent replies come back as agent. Example: GET /wp-json/mxchat/v1/transcripts?role=assistant&knowledge_used=no&since=2026-05-01&limit=50.

For a Model Context Protocol (MCP) interface to your install — so Claude, ChatGPT, and other agents can talk to MxChat natively — see the MxChat MCP Server add-on.

For developers (3.2.22). Add-ons can put a button in the chat toolbar with the mxchat_chat_toolbar_items filter, and the mxchat:toolbar-ready and mxchat:opened page events replace watching the page or running a request on every view — the WooCommerce cart button is the first to use it. Add-ons can also append to a chat transcript with save_session_message, shape what humans read with mxchat_transcript_user_message and mxchat_ai_conversation_history, and add sections to the Telegram and Slack handoff card with mxchat_handoff_card_sections. All of these are documented on the developer hooks & filters page.

For developers (3.2.26). mxchat_generate_media_image() lets an add-on generate an image with the site's image model and save it to the Media Library (it returns the attachment id, or a WP_Error). New filters: mxchat_should_index_post, mxchat_index_use_rendered_page, mxchat_transcripts_capability, mxchat_claude_max_tokens, mxchat_openai_image_model and mxchat_gemini_image_model, each described in its section above.

For developers (3.2.27). mxchat_stream_request changes a streamed chat request before it is sent. It receives the request (url, headers as Name: value strings, and body as the decoded JSON), the provider (openrouter, openai, openai_responses, custom, claude, xai or deepseek), the model id and the bot id. Also new: mxchat_filter_roles_rate_limit (see Message limits), mxchat_lead_retention_days (see Delete Leads After) and mxchat_xai_image_model (see Generate Image).

Troubleshooting

Security update — 3.2.27. This release includes security hardening and general security improvements. Update recommended.

Security update — 3.2.26. This release hardens how chat messages, code, visitor feedback, translations and import details are displayed, how PDF links in the chat are opened, and the permission checks on several admin requests. The pre-chat form and its consent checkbox are now enforced on the server, and message limits are applied more strictly. Update recommended.

Security update — 3.2.16. This release hardened how MxChat validates chat session identifiers across the plugin, and tightened the permission checks around its admin scripts and admin-only actions so they load and run only for users who can manage plugin options. If your site sends automatic transcript emails, update.

"Invalid API key" / no replies — re-check the key under Settings → API Keys; make sure it matches the provider of your selected chat model.
Bot doesn't know my content — confirm the content is indexed under Settings → Knowledge, and that an embedding model + key are configured. Lower the Similarity Threshold if good content isn't being matched.
"Access denied" on the first message behind a cache — make sure you're on the current version; MxChat fetches a per-request nonce that resolves this for WP Rocket, LiteSpeed, FlyingPress, W3 Total Cache, and Cloudflare APO.
A model shows an error after a provider change — MxChat moves a discontinued model to its replacement and shows a notice naming both (from 3.2.26 also for models with an announced shutdown date, on that date), and each chat request already uses the replacement; if a specific model still errors, pick a current one under Settings → Chatbot → AI Models.
An occasional dropped or failed reply — MxChat automatically retries on transient provider errors (5xx, rate-limit, and overload responses) for both streaming and standard chat, so brief provider hiccups usually recover on their own without a visible error.
A visitor sees an AI error in the chat — when a call still fails after retries (a bad key, a hard rate-limit, a connection timeout, or a provider outage), MxChat shows a short, readable message in the chat rather than a raw error dump, and never exposes internal details or your API key. The specifics are in your server log for you to diagnose. Since 3.2.15 those messages carry the provider's own explanation — including the exact model name xAI could not find — instead of a generic "Unknown error".
An empty chat bubble — when a provider returns a successful but empty reply, MxChat (since 3.2.15) shows a clear message asking the visitor to try again instead of an empty bubble. Streaming replies close out the same way.
A knowledge base import failed and the error didn't say why — fixed in 3.2.16. Import errors now carry the embedding provider's own reason (an invalid API key, exhausted credit, a model your key can't use), name the model, and say which stage failed. In particular, Failed to store any chunks now continues with either "every chunk failed to embed" or "embeddings generated but storage failed" — the first points at your embedding API key, the second at your Pinecone index or database. On earlier versions both cases produced the same sentence, which is why a bad key so often sent people to rebuild a Pinecone index that was fine.
Streaming replies arrive all at once after a pause — fixed in 3.2.16 for sites behind a buffering proxy such as nginx; replies stream word by word again, and four PHP warnings per affected chat go with it.
Quick Question buttons render too small — fixed in 3.2.16. In 3.2.15 the new theme-protection styles let the list squeeze its buttons to fit the box instead of scrolling; they are back to their normal size with the scrollbar restored.
A Loops setting the message told you to find isn't on that screen — fixed in 3.2.16. Both email-capture messages now name the API Keys tab and link straight to it.
Japanese, Chinese or Thai content came back empty or mangled after an import — fixed in 3.2.17. Content in scripts without spaces between words is no longer erased, split into oversized chunks, or stripped of rare characters on the way into the knowledge base. Re-import anything you indexed before updating.
A toggle saved but never showed the "saved" confirmation — fixed in 3.2.17 for the hybrid keyword boost and smart asset loading toggles. The setting was being stored correctly; only the confirmation was missing, which made both look broken.
The WordPress content import window was cut off on a phone — fixed in 3.2.17; the import dialog now fits small screens with its buttons reachable.
The bot answers as if the knowledge base were empty, and you use OpenAI Vector Store — fixed in 3.2.18. Every Vector Store search was rejected by OpenAI and the error was discarded, so answers came back with no retrieved content and nothing was logged. Update; no re-index is needed. See the note above.
A sitemap import failed and the status area was blank — fixed in 3.2.18; sitemap imports now name the reason, including an API key that lacks the permission the import needs. Knowledge base import errors from OpenAI, Gemini and Voyage also pass along the provider's own explanation instead of a generic invalid-key message.
A PDF shared in chat came back "empty or unsupported" — fixed in 3.2.18. When an embedding request failed part-way through a document, the unreadable pages were stored anyway; they are now skipped and recorded in the Debug Mode log, and a file where no page can be read reports the real reason instead of claiming the file is empty.
A tick mark in the content selector was invisible — fixed in 3.2.18; selected items were drawing white on white and now show a filled checkbox.
Transcripts labelled everything "AI Assistant" — fixed in 3.2.15. Conversations a human took over now show as Live Agent in the Transcripts viewer and in emailed transcripts, so you can tell which replies came from your team and which came from the bot. Nothing to configure; update to 3.2.15 or later.
Hebrew, Arabic or other right-to-left PDFs imported with the text reversed — fixed in 3.2.19, and a new WP-CLI repair command fixes entries imported before the update in place, so you do not have to re-import them.
Notification emails or transcripts stopped arriving after you entered two addresses — fixed in 3.2.19, and this one failed silently, so it is worth checking. On earlier versions the Notification Email field held one address; entering [email protected], [email protected] was stored as [email protected], which passed WordPress's own validity check, so nothing looked wrong and every alert went to a domain that does not exist. From 3.2.19 the field takes up to five comma-separated addresses, refuses an invalid one with a message naming it, and never silently accepts a mangled list.
A YouTube card appeared under an answer it had nothing to do with — fixed in 3.2.19. A video card used to attach whenever a matching source was retrieved, with no relevance gate. There is now a match threshold setting, and video cards can be switched off entirely.
The chatbot quotes product prices in the wrong currency — fixed in 3.2.19 for newly indexed content. WooCommerce prices are now indexed in your store's base currency with the currency code, and the bot no longer mixes two currencies in one reply. Products indexed before the update keep the old text, so re-run the WordPress Content import for your products after updating.
You lost settings edits by navigating away mid-change — improved in 3.2.19. Changed fields show an Unsaved marker, leaving mid-edit asks for confirmation, and the edits are handed to the browser for saving if you leave anyway.
Imported content lost currency symbols, or Arabic/Hebrew/Hindi/Thai marks — fixed in 3.2.19, along with leftover HTML character codes now being converted to readable text before indexing. Re-import the affected content to clean up existing entries.
A draft post was indexed under your homepage address — fixed in 3.2.19 for drafts published by import tools or scripts.
A Telegram agent can read a returning visitor's messages but cannot reply — fixed in 3.2.21. Ending a conversation used to leave the topic mapped, so that visitor's next request posted into the closed topic, which Telegram gives no compose box. Ending a conversation, from MxChat or from Telegram, now retires the topic and the next request opens a fresh one.
A Telegram handoff arrived with no conversation history, or not at all — fixed in 3.2.21. A long conversation could exceed Telegram's message limit and take the whole handoff card with it; the history is now sent in parts as readable text, and any delivery Telegram refuses is shown on the Telegram settings screen with its reason.
Telegram replies stopped arriving after you put the site behind Cloudflare — the webhook was registered without a secret, so every request is rejected as coming from outside Telegram's ranges. From 3.2.21 the Telegram settings screen and a dashboard notice count the rejections and say why; set a webhook secret and re-register the webhook from that screen.
A Slack agent replied in a thread and the visitor never saw it — fixed in 3.2.21 for per-conversation channels. In a shared channel, replies must always go inside the conversation's thread.
Start new chat left one old reply on screen and no greeting — fixed in 3.2.21 for sites with Chat Persistence on.
The Pro & Extensions button for an add-on you have not installed opened the same page as Documentation — changed in 3.2.21: it now reads Get the ZIP from My Account and opens your account's Downloads page, where the add-on ZIPs live.
The chatbot recommends past, expired events — fixed in 3.2.22: the current date and time in your site's timezone now travel with every message (see Behavior above). On earlier versions nothing told the model the date.
A knowledge entry restricted to a custom role was readable by every signed-in visitor — fixed in 3.2.22. Restrictions to the five built-in roles were never affected.
Choosing Media in the WordPress Content importer showed nothing — fixed in 3.2.22; it now lists the PDF, Word, text and Markdown files in your library.
With Pinecone, Knowledge Entries says (50) however much you import — fixed in 3.2.22; the heading now shows Pinecone's own vector count. Nothing was ever missing from the index.
A video or audio player on the page starts playing when a visitor sends a message — fixed in 3.2.22 for sites with Contextual Awareness on; the page text is read from an inert copy, gathered once per page.
Underscored words — file names, SKUs, _PLACEHOLDER_ tokens — lost their underscores and turned italic — fixed in 3.2.22 (see Markdown above).
Daily, weekly or monthly message limits never seemed to bite — fixed in 3.2.22; counters used to be cleared hourly and keyed on a client-supplied address header.
A tool-using conversation ended in a provider error once several tools had run — fixed in 3.2.22; the function-calling loop no longer sends an invalid request when the per-turn tool budget is reached.
Pressing Enter (or Send) on an empty chat box locked the input until the page was reloaded — fixed in 3.2.23; only a real message disables the box while the reply is generated, and a message of only spaces or blank lines counts as empty.
Create index for me failed with HTTP 409 — fixed in 3.2.23; the Index Name field still held the vector index's name. Switching to Document index now suggests a free -docs name.
A part number typed without its spaces or hyphens found the wrong record — fixed in 3.2.23 on a Pinecone document index; run Migrate again for records copied on 3.2.22.
Links in chatbot replies did nothing in Safari with Open Links in New Tab on — fixed in 3.2.24. The tab used to be opened after the click was recorded, outside the visitor's tap, and Safari's pop-up blocker dropped it silently.
A Slack live-agent message failed but was reported as sent — fixed in 3.2.24. When Slack refuses a message (deleted or archived channel, revoked token) the visitor is told, the handoff ends so the chatbot takes over again, and the error shows on the Slack tab. Slack messages no longer begin with stray characters.
The video card was missing on streamed GPT-5 and GPT-6 answers — fixed in 3.2.24; it used to need AI Tools on or streaming off.
Answers showed an uploaded file's name as a link that went nowhere — fixed in 3.2.24, including files uploaded earlier: the file is named in plain text, or linked to its Source URL when you give one.
On a Pinecone document index, codes that start with digits (5872EF, 4471AB12) were not found — fixed in 3.2.24. Entries written earlier keep their old codes: run Migrate again or re-save the entry.
Custom Provider embeddings: POST /knowledge asked for an OpenAI key, and a Pinecone index was sized for the wrong dimension — fixed in 3.2.24. The REST API needs no OpenAI key, and the index is sized from the custom model's real dimension; the Knowledge listing, search, action phrases and import notice follow the custom model too.
View indexed content failed for some entries — fixed in 3.2.24 for addresses with encoded characters (non-Latin pages, spaces) and manually added Pinecone entries; long entries now show in full.
A media file whose knowledge entry was deleted was skipped as unchanged on the next save — fixed in 3.2.24; it is indexed again.
The WordPress personal data export missed visitors who filled in the pre-chat form without chatting — fixed in 3.2.24; their email, name and consent record are included.
Delete old index could strand Multi-Bot bots — improved in 3.2.24: it refuses while bots still use the old index, and names them.
A Gemini chatbot stopped with "The response format was unexpected" on an ordinary question — fixed in 3.2.25. Google's safety filter had withheld the reply; MxChat now asks it to block only high-risk content (see Gemini safety filter above).
A visitor's message with a web link and a word like "read" or "documentation" got a PDF error — fixed in 3.2.25; such a message is no longer treated as a PDF to read.
Code blocks or other elements on your own pages turned black after installing MxChat — fixed in 3.2.25; the chat styles no longer reach outside the chat window.
A page with an embedded chatbot also showed a floating bubble that could not send messages — fixed in 3.2.25; the floating bubble of the same bot is no longer added to that page.
A www link at the end of a paragraph, or a bold web address, came out broken in the chat — fixed in 3.2.25.
Text from a synced pattern or a WoodMart HTML Block was missing from a page's knowledge entry — fixed in 3.2.25. Save or import the page again.
Saving a product removed its Custom Product Tabs text from the chatbot's knowledge — fixed in 3.2.25. Re-save or re-import products saved on earlier versions.
Brave web or image search on a Gemini 3 model searched for the visitor's whole sentence — fixed in 3.2.25; the query is cut down to key words first.
The Knowledge list's PDF filter showed nothing, and uploaded files appeared as a View Source link that led nowhere — fixed in 3.2.25.
Opening the Leads tab wrote a notice to the debug log — fixed in 3.2.25 for sites with WordPress debugging on.
Every MxChat settings save cleared the whole page cache on a large site — from 3.2.25, turn off Clear page cache when MxChat settings change under Settings → Optimization → Page Cache Compatibility (see above).
The chatbot only appears after the visitor scrolls or clicks — fixed in 3.2.26: the launcher now appears straight away even when a cache plugin's Delay JavaScript holds back chat-script.js. The first tap opens the chat only once that script has run; exclude it from the delay if you want the first tap to work.
A page built with Breakdance, Oxygen, Bricks, Divi or WPBakery, or with ACF Blocks, had little or no text in the knowledge base — fixed in 3.2.26. Run Re-import affected entries or import the page again.
Replies stopped mid-sentence — fixed in 3.2.26; a reply that reaches the length limit now ends with a short note, also with AI Tools.
The chatbot stopped working when the Moderation add-on was active without a Pro licence — fixed in 3.2.26. Moderation bans now also match visitors behind Cloudflare and guests who gave a banned email address.
Image generation in the chat failed with Gemini — fixed in 3.2.26 (see Generate Image under Actions).
The chat ignored Complianz consent with delayed or on-interaction script loading — fixed in 3.2.26.
Grok web and image searches were slow to start — improved in 3.2.26; they now search for the key words of the question.
Code blocks lost part of their code, or Copy copied something different — fixed in 3.2.26. The same release makes the Leads list readable on laptop screens and code block labels easier to read on light themes, and the auto-retry switch now stays off when turned off.
A "Switching off" or "MxChat Pro is paused" notice appeared — the licence holder switched this site off in their mxchat.ai account (3.2.25). Only they can switch it back on; see MxChat Pro on each site above.
A chat placed in a page opened before the visitor accepted cookies — fixed in 3.2.27 for sites with the Complianz Integration on. It now waits for consent like the floating chat, and withdrawing consent closes it (see Complianz cookie consent above).
With delayed script loading, the first click on the launcher did nothing — fixed in 3.2.27 for MxChat's own Delay and On User Interaction strategies; the click is remembered and opens the chat once the script loads.
The first open of the chat showed a Too Many Requests error — fixed in 3.2.27.
Saving settings changed your chat model or embedding model — fixed in 3.2.27; an unrecognised chat model or an older embedding model is kept as saved.
o-series models, Custom Provider streaming or web search failed on newer models — fixed in 3.2.27.
Transcripts showed stray code, empty messages, or a reply split across several messages — fixed in 3.2.27; replies show as the visitor saw them, in Transcripts, the reloaded chat and the transcript email.
Knowledge base sync missed synced patterns, PDFs in ACF Blocks, or posts that were unpublished, trashed or restored — fixed in 3.2.27.
Breakdance, Oxygen, Bricks or Divi 5 pages showed little or no text in the import list — fixed in 3.2.27; the list shows the published text the import stores.
Replies had extra blank lines — fixed in 3.2.27.
An add-to-cart request ended in a critical error — fixed in 3.2.27.
The Actions page saved trigger phrases that could never match — fixed in 3.2.27; it now reports the embedding error instead.
WordPress notices covered the MxChat header and menu, or admin pages were laid out badly on a phone, tablet or wide screen — fixed in 3.2.27.
Minimum versions — 3.2.22 declares WordPress 5.3 and PHP 7.4 as its minimums, matching what the code has required for some time; every MxChat add-on already required PHP 7.4.

Still stuck? Ask the Documentation Bot — it has access to the full MxChat source and every plugin's docs — or open a ticket.

Last reviewed by Sage on 2026-10-25 (MxChat core 3.2.27) — documented Delete Leads After on the Transcript Management card and mxchat_lead_retention_days, ratings and link clicks deleted with a lead or chat, Generate Image (xAI) with Grok Imagine 2.0 and mxchat_xai_image_model, the knowledge_used filter and the sources_used / knowledge_used fields on GET /transcripts and role=assistant returning replies, Complianz consent for chats placed in a page and consent withdrawal, retrieval/search tokens and replies as the visitor saw them in Transcripts, one embedding call per reply, settings saves that keep your models, the first click with MxChat's delayed script loading, mxchat_filter_roles_rate_limit and mxchat_stream_request, the sync fixes, the 3.2.27 security update and fixes under Troubleshooting. Previously reviewed 2026-10-18 (MxChat core 3.2.26) — documented the new models (GPT-6.1 Sol, Claude Haiku, Sonnet and Opus 5.5, Claude Fable 5.1, Grok 4.7, Gemini 4 Argon, DeepSeek Flash, Gemini Embedding 2), the fast and affordable Recommended model per provider, the current / older / dated model lifecycle with automatic replacement on the shutdown date and embedding models never switched, page-builder and ACF Blocks indexing, the Keep imported entries up to date only Auto-Sync switch, mxchat_should_index_post, token usage and the mxchat_view_transcripts capability in Transcripts, the reply-length note and larger Claude budget, GPT Image 2.5 Flare and Nano Banana 2.1 for in-chat images, the launcher on cached and delayed pages, Complianz with delayed loading, mxchat_generate_media_image(), the 3.2.26 security update and fixes under Troubleshooting. Previously reviewed 2026-10-11 (MxChat core 3.2.25) — documented the per-site Pro switch on the mxchat.ai Licenses page (daily check, three-day grace, Check license now), the Clear page cache when MxChat settings change switch and the Delay JavaScript note on Settings → Optimization, the edited-after-import option for Re-import affected entries and --edited, the Gemini safety filter and mxchat_gemini_safety_settings, Gemini 2.5 marked no longer offered, retired models replaced on every request, synced patterns, WoodMart HTML Blocks and Custom Product Tabs in imports, and the 3.2.25 fixes under Troubleshooting. Previously reviewed 2026-10-03 (MxChat core 3.2.24) — documented the Source URL field and PDF Upload changes, Re-import affected entries and wp mxchat kb-reimport, product tags in the knowledge base, the Multi-Bot shared-store notice, and the 3.2.24 fixes under Troubleshooting. Previously reviewed 2026-09-30 (MxChat core 3.2.23) — documented the GPT-6 Astra, Sol and Luna models, the Pinecone Migrate changes (copies while the chatbot keeps answering, Switch anyway, Copy every namespace, every-namespace check before Delete old index, the -docs name suggestion that ends the HTTP 409), part numbers matched without their separators, the tool-call budget the model is now told about, and the empty-Enter input lock fix. Previously reviewed 2026-09-21 (MxChat core 3.2.22) — documented the opt-in Pinecone document index with full-text search (Create index for me, Check index, copy-migration, Hybrid keyword boost on Pinecone, Vector/Keyword/Both labels in Testing), the automatic current date and time line and the {current_date} / {current_datetime} placeholders, role restrictions for custom roles and the per-page Knowledge base access setting, Media Library import and auto-sync of PDF, Word, text and Markdown files, the pre-chat form for logged-in visitors and the optional email field, DELETE /knowledge by source URL and GET /leads, the Gemini 3.8/3.7/3.6 Flash models and the low thinking level, the toolbar and transcript developer hooks, and troubleshooting for the custom-role leak, the empty Media importer, the Pinecone (50) count, autoplaying media under Contextual Awareness, underscores turning italic, hourly-resetting rate limits and the function-calling budget error. Previously reviewed 2026-09-12 (MxChat core 3.2.21) — documented where Slack agents must reply in each channel mode now that thread replies in per-conversation channels are relayed, the note agents see when an edited or file-based reply is dropped, the private-shared-channel warning and the message.groups subscription it asks for, the mxchat_handoff_history_count filter, the Telegram changes (ten-message history sent in parts, closed topics retired on #close or when closed from Telegram, delivery failures and webhook rejections surfaced on the Telegram settings screen), the Chat Persistence greeting fix for Start new chat, and the Pro & Extensions button that now opens your Downloads page. Previously reviewed 2026-08-29 (MxChat core 3.2.20) — documented the new Strip Unapproved Links setting under Settings, Behavior: the link guard is now its own control instead of riding Citation Links, so fabricated URLs are removed even with citations off, while approved citations, real pages on this site and integration-returned links such as WooCommerce product URLs are always kept. Previously reviewed 2026-08-16 (MxChat core 3.2.19) — documented the daily model-liveness check and the automatic migration off retired Claude models, automatic Anthropic prompt caching, the refreshed xAI Grok lineup, AI Tool runs now recorded per message in Transcripts (sensitive tools logging only that they fired), the live-agent handoff tools shipping with their usage hint prefilled, what the Slack handoff actually sends an agent, and troubleshooting for right-to-left PDF imports, the multi-address Notification Email field, mismatched YouTube cards, WooCommerce currency indexing, unsaved-settings protection and homepage-addressed drafts. Previously reviewed 2026-08-09 (MxChat core 3.2.18) — documented that OpenAI Vector Store retrieval was failing silently on earlier versions and is fixed (no re-index needed), the new per-conversation mxchat_sessions table with automatic 30-day cleanup that works even where WP-Cron never runs and its mxchat_session_retention_days filter, and troubleshooting for blank sitemap-import errors, part-way PDF embedding failures and the invisible content-selector tick. Previously reviewed 2026-08-06 (3.2.17) — documented that custom and local embedding providers such as Ollama and LM Studio now work end to end with no cloud API key, that the hybrid keyword boost moved to Knowledge → Chunking & Retrieval, that scheduled posts are indexed when they go live, and troubleshooting for Japanese/Chinese/Thai imports, the missing toggle save confirmation and the cut-off import window. Every settings tab also now has its own shareable link that opens directly on that tab. Previously reviewed 2026-08-03 (3.2.16) — added the Privacy & Data Requests section covering MxChat's integration with WordPress's export and erasure tools, the hybrid keyword boost for the WordPress-database knowledge base, the wp mxchat prune-unpublished command and reliable removal of unpublished content, retirement dates in the model picker plus the GPT-5.1 Chat Latest / GPT-5.3 Chat migration to GPT-5.6 Sol on August 10, the persistent admin notice when a provider rejects your model, the plainer "nothing relevant in the knowledge base" handling, and troubleshooting for import errors that now name the reason and the failing stage, proxy-buffered streaming, Quick Question button sizing and the Loops message wording. Spotted an issue? Open a ticket and we'll patch the doc.