Muhammad Adnan Hanif

Fast RAGbot is an intelligent chatbot that uses Retrieval-Augmented Generation (RAG) with support for multiple AI providers to provide accurate answers based exclusively on your website content.
Key Features:
<ul>
<li>Multi-LLM Support: Choose between Gemini, OpenAI (ChatGPT), or Claude</li>
<li>Multi-Domain Indexing: Index up to 3 different domains</li>
<li>Document Upload: Bulk upload PDF, DOCX, PPTX, XLSX, TXT, XML, CSV</li>
<li>Conversation Memory: Maintains last 5 Q&A turns per session</li>
<li>Smart Follow-ups: Suggests questions when answer not found</li>
<li>Widget Customization: Custom colors, logo, position, theme CSS</li>
<li>Pre-chat Form: Optional name/email collection</li>
<li>Chat Logs: Full conversation logging with CSV export</li>
<li>Shortcode Support: Embed chat anywhere with <code>[ragbot_chat]</code></li>
<li>Grounded Responses: Answers only from your content – no hallucinations</li>
<li>Source Citations: Provides URLs of source pages for transparency</li>
<li>Mobile Responsive: Works perfectly on all devices</li>
<li>Security First: Rate limiting, nonces, and proper sanitization</li>
</ul>
How It Works:
<ol>
<li>Crawls and indexes your website content</li>
<li>Creates semantic embeddings using Google Gemini</li>
<li>Retrieves relevant content based on user questions</li>
<li>Generates contextual answers with source citations</li>
<li>Displays responses in a floating chat widget</li>
</ol>
Perfect For:
<ul>
<li>Customer support automation</li>
<li>FAQ assistance</li>
<li>Product information queries</li>
<li>Company information lookup</li>
<li>Documentation assistance</li>
</ul>
<h3>External Services & Data Usage</h3>
This plugin connects to external AI services to provide chatbot functionality. Below is a detailed explanation of what data is transmitted and how it is used.
<h4>External Services Used</h4>
Google Gemini API (https://ai.google.dev/) 
* Used for: Text embeddings and AI text generation 
* Data sent: User questions, website content chunks, conversation history 
* Privacy Policy: https://policies.google.com/privacy 
* Terms of Service: https://policies.google.com/terms
OpenAI API (https://openai.com/) – Optional 
* Used for: Text embeddings and AI text generation (when selected as active model) 
* Data sent: User questions, website content chunks, conversation history 
* Privacy Policy: https://openai.com/privacy/ 
* Terms of Service: https://openai.com/terms/
Claude API by Anthropic (https://anthropic.com/) – Optional 
* Used for: Text embeddings and AI text generation (when selected as active model) 
* Data sent: User questions, website content chunks, conversation history 
* Privacy Policy: https://www.anthropic.com/privacy 
* Terms of Service: https://www.anthropic.com/legal/terms
<h4>What Data Is Transmitted</h4>
<ol>
<li>User’s Question: The text message submitted by the website visitor</li>
<li>Retrieved Document Chunks: Relevant excerpts from your indexed website content (typically 5-8 chunks of ~1000 characters each)</li>
<li>Conversation History: Last 5 question-answer pairs from the current session</li>
<li>Model Parameters: Temperature (0.1), max tokens (1000), and other generation settings</li>
</ol>
<h4>Data Usage & Storage</h4>
<ul>
<li>Data is transmitted only for generating responses to user questions</li>
<li>No data is stored or used for training by external AI providers (per their API terms)</li>
<li>Data transmission occurs in real-time and is not cached by the plugin on external servers</li>
<li>All API communications use HTTPS encryption</li>
</ul>
<h4>Local WordPress Storage</h4>
The plugin stores the following data in your WordPress database:
Options (wp_options table): 
* <code>ragbot_settings</code> – Plugin configuration including API keys, widget settings, and preferences
Custom Tables: 
* <code>wp_rag_docs</code> – Indexed website pages and uploaded documents 
* <code>wp_rag_chunks</code> – Text chunks created from indexed content 
* <code>wp_rag_embeddings</code> – Vector embeddings for semantic search 
* <code>wp_rag_sources</code> – Content source configurations (domains, settings) 
* <code>wp_rag_logs</code> – Chat conversation logs (optional, includes user name/email if pre-chat form enabled) 
* <code>wp_rag_url_queue</code> – Queue of URLs pending indexing
Session Data: 
* PHP sessions store temporary conversation history (cleared when session ends) 
* Browser sessionStorage stores chat UI state (cleared when tab closes)
<h4>Data Control & Privacy</h4>
User Control: 
* Delete API keys anytime from Settings > Fast RAGbot > Models tab 
* Clear all indexed content using “Clear Index” button 
* Reset plugin settings to defaults 
* Export or delete chat logs from Settings > Fast RAGbot > Chat Logs tab
Data Deletion: 
* Deactivating the plugin preserves your data 
* Uninstalling the plugin removes all database tables and options 
* Use “Clear Index” to remove indexed content without uninstalling
Privacy Compliance: 
* No data is shared with third parties except your chosen AI provider 
* Chat logs are optional and stored locally in your database 
* Pre-chat form (name/email collection) is optional and disabled by default 
* All user inputs are sanitized and validated 
* Rate limiting prevents abuse (10 requests per 60 seconds per IP)
<h4>Developer Privacy Policy</h4>
For questions about data handling, please review: 
* Your chosen AI provider’s privacy policy (links above) 
* Your own website’s privacy policy (you are the data controller) 
* WordPress.org privacy guidelines: https://wordpress.org/about/privacy/
<h3>Privacy Policy</h3>
This plugin acts as a data processor, sending user questions to your chosen AI provider. You (the site owner) are the data controller and responsible for: 
* Disclosing AI service usage in your privacy policy 
* Obtaining user consent where required by law (GDPR, CCPA, etc.) 
* Ensuring compliance with local privacy regulations
The plugin stores chat logs locally in your database. If you enable the pre-chat form, user names and emails are stored and should be disclosed in your privacy policy.
<h3>Support</h3>
For support, please visit the plugin’s support forum or contact the developer.
<h3>Credits</h3>
<ul>
<li>Supports embeddings and text generation using Google Gemini, OpenAI, or Claude.</li>
<li>Built with WordPress coding standards</li>
<li>Follows accessibility best practices</li>
</ul>

Fast RAGbot

AI-powered chatbot with multi-LLM support (Gemini/OpenAI/Claude), multi-domain indexing, document upload, and conversation memory.

Selvakumar Duraipandian

Markdown for AI Agents is a lightweight WordPress plugin that enables HTTP content negotiation for your site’s content. When a client (like an AI agent or a custom script) requests a page with the <code>Accept: text/markdown</code> header, the plugin intercepts the request and returns a clean, structured Markdown representation of the post or page content.
This is ideal for AI crawlers, RAG (Retrieval-Augmented Generation) systems, and non-browser clients that prefer machine-friendly text over complex HTML.
Important note: This plugin is primarily a developer/integration tool. Human visitors browsing your site will never see any difference — the Markdown output is only served when explicitly requested via the <code>Accept: text/markdown</code> HTTP header. Normal browser requests always receive the standard HTML page.
Key Features:
<ul>
<li>Automatically detects <code>Accept: text/markdown</code> headers.</li>
<li>Converts HTML content to clean Markdown using the League HTMLToMarkdown library.</li>
<li>Strips away theme layout, navigation, headers, footers, and sidebars — serving only the main content.</li>
<li>Adds useful HTTP response headers: <code>Content-Type: text/markdown</code>, <code>Vary: Accept</code>, and <code>X-Markdown-Word-Count</code>.</li>
<li>Respects WordPress visibility rules and filters.</li>
<li>No configuration required — works out of the box for posts, pages, and custom post types.</li>
</ul>
<h3>How It Works</h3>
This plugin uses a standard web technique called HTTP content negotiation. The same URL on your site can serve different representations of the same content depending on what the client asks for:
<ul>
<li>A regular browser sends <code>Accept: text/html</code> → receives your normal HTML page.</li>
<li>An AI agent sends <code>Accept: text/markdown</code> → receives a clean Markdown version of the same page.</li>
</ul>
No extra URLs, no duplicate content, no configuration needed. The plugin hooks into WordPress’s <code>template_redirect</code> action, detects the <code>Accept</code> header, captures the rendered HTML, converts it to Markdown, and returns it with appropriate headers.
<h3>Why Markdown for AI Agents?</h3>
When building RAG (Retrieval-Augmented Generation) applications or AI pipelines that ingest web content, HTML is extremely noisy. A typical WordPress page contains thousands of tokens worth of HTML tags, inline styles, navigation menus, scripts, and layout markup — none of which carries meaning for an AI model.
Serving clean Markdown instead can reduce token consumption by up to 60%, which means:
<ul>
<li>Lower API costs — fewer tokens ingested when loading pages into vector stores or LLM pipelines.</li>
<li>Faster processing — less text for the model to parse, filter, and discard.</li>
<li>Better retrieval accuracy — higher signal-to-noise ratio improves the quality of RAG results.</li>
<li>Simpler pipelines — no need for custom HTML stripping logic on the client side; the plugin handles it server-side.</li>
</ul>
Any AI agent, crawler, or ingestion script that sends <code>Accept: text/markdown</code> in its request header will automatically receive the clean Markdown version — no extra URLs, no separate endpoints, no changes to your content workflow.

Markdown for AI Agents

Serve clean Markdown versions of WordPress content to AI agents using HTTP content negotiation.

Fast RAGbot vs Markdown for AI Agents