mad.Design AI Sales and Support Chatbot
| 开发者 |
maddesigngmbh
heikoxmaddesign |
|---|---|
| 更新时间 | 2026年10月2日 22:41 |
| PHP版本: | 8.2 及以上 |
| WordPress版本: | 7.1 |
| 版权: | GPLv2 or later |
| 版权网址: | 版权信息 |
详情介绍:
- Indexes the post types you choose into a vector store on your own server, either as HNSW index files or in a native MariaDB
VECTORcolumn. - Hybrid retrieval: semantic search and keyword full-text search run together and the rankings are merged. The right page is found whether the visitor uses your wording or their own.
- Contextual retrieval: every chunk gets a one-line summary at index time that names its page and section, so short passages are still found.
- Page context: attach notes to a single page that the chatbot sees whenever a visitor is on that page.
- Page builders are indexed with their rendered output, not raw shortcodes: Elementor, Bricks, Divi, Beaver Builder, Brizy, Thrive Architect, SiteOrigin and WPBakery, with auto-detection per post.
search_knowledge_base: semantic and keyword search across your indexed contentquery_posts,get_post,get_page_content: look up live posts and pageslist_post_types,get_taxonomy_terms,get_posts_by_term: browse your site structure
- Prompt caching: the fixed part of the prompt stays identical between turns, so providers that support caching charge less and answer faster.
- Long conversations are condensed into a running summary, so the prompt does not grow with every message.
- Replies are generated in a background job. A slow provider never blocks your page.
- Performance logging shows the time spent per stage of each reply.
- Optional GDPR consent overlay before the first message.
- Email addresses and phone numbers are masked before a message reaches the AI provider. The originals stay in your database for follow-up.
- Indexed content and embeddings stay on your server. Only the passages needed for the current question are sent.
- Per-IP rate limiting and retention periods for conversations and logs.
- API keys can be stored in
wp-config.phpinstead of the database. - Conversations are readable in the admin, with optional email notification.
安装:
- In your WordPress admin go to Plugins → Add New Plugin, search for "mad.Design AI Sales and Support Chatbot", click Install Now, then Activate.
- Go to Chatbot → Settings → AI Engine, choose a provider and enter your API key.
- Under Knowledge, enter your embeddings endpoint and API key (OpenAI-compatible, defaults to OpenAI), choose the post types to index and run the indexer.
- Under Content & Behavior and Appearance, set up the widget. Then enable "Add to all pages" or place the
[mad_chatbot]shortcode where you want it.
[mad_chatbot mode="embed"].屏幕截图:
常见问题:
Do I need an API key?
Yes. You supply your own key for the chat provider you choose.
Indexing needs a second key for the embeddings endpoint, set under Knowledge. It can be the same vendor or a different one.
Both keys can be defined in wp-config.php as MADCB_MODEL_API_KEY and MADCB_EMBEDDING_API_KEY instead of being stored in the database.
Which AI providers are supported?
OpenAI (ChatGPT), Anthropic (Claude), Google (Gemini), and any OpenAI-compatible endpoint through the "Custom" provider, for example OpenRouter, Azure OpenAI, vLLM, Ollama or LM Studio.
Will it make things up?
It is instructed to answer from the retrieved content and to say so when it finds nothing. Grounding reduces wrong answers but does not rule them out. Check the conversation log before relying on it for anything critical.
Does it work with page builders?
Yes. Select your builder under Knowledge, or leave it on auto-detect. Posts are indexed with their rendered content instead of raw shortcodes.
Can I use it for lead generation?
Yes. When a visitor leaves an email address or phone number, the conversation is stored in WordPress and emailed to your team.
Is it GDPR-friendly?
A consent overlay can require agreement before the first message, and retention periods for conversations and logs are configurable. Messages are still processed by a third-party AI provider, which you need to disclose in your privacy policy.
Can developers extend it?
Yes. Tools, AI providers and context sources are registered through filters. Prompt templates can be overridden from your theme, and the widget templates are Twig files.
更新日志:
- Security fixes. Update recommended.
- New setting: choose who can read conversations.
- AI notice under the message field for the EU AI Act, on by default.
- Any model ID can be entered with "Other…", for every provider.
- Updated model lists: Claude Fable 5.1, Opus 5.5, GPT-6, Gemini 3.8 to 3.5.
- Chat title defaults to "Chatbot".
- Twig updated to 3.30.
- Initial release.
- Chat providers: OpenAI (ChatGPT), Anthropic (Claude), Google (Gemini) and OpenAI-compatible endpoints.
- Retrieval from an HNSW index or a MariaDB
VECTORcolumn, with hybrid semantic and keyword search. - Contextual retrieval with per-chunk summaries created at index time.
- On-demand retrieval: the model searches only when it needs to, with always-on retrieval as an option.
- Seven built-in tools, each can be disabled, plus a filter for custom tools.
- Prompt caching and running conversation summaries to keep cost and latency down.
- Three widget templates with WCAG AA colour presets.
- Page builder support: Elementor, Bricks, Divi, Beaver Builder, Brizy, Thrive Architect, SiteOrigin, WPBakery.
- Rate limiting, retention periods, GDPR consent overlay, email notifications, performance logging.
- Multisite-aware install and uninstall.