Nguyễn Văn Thụ

NVT AI Chatbot connects your WordPress site to the NVT SaaS platform to deliver an AI-first support experience for WordPress and WooCommerce stores.
The plugin focuses on three core jobs:
<ul>
<li>Answer with your own knowledge base: Use Retrieval-Augmented Generation (RAG) so the chatbot answers from your website content and uploaded documents.</li>
<li>Escalate to a human when needed: When AI is not enough, visitors can switch to Live Chat and your team can continue the conversation from the built-in Inbox.</li>
<li>Support WooCommerce sales and support flows: Help customers discover products, manage cart actions, move toward checkout, and ask about recent orders.</li>
</ul>
Main features:
<ul>
<li>AI Chatbot (RAG): Grounded answers based on your website content and uploaded files.</li>
<li>Website Crawl + File Ingestion: Crawl your site manually or on a schedule, or upload files such as PDF, DOCX, TXT, MD, HTML, CSV, XLS, XLSX, and PPTX.</li>
<li>Live Agent Handoff: Switch between AI and Live Chat using schedule rules, manual override, or user request.</li>
<li>WooCommerce Product Support: Search products, recommend items, and show product cards inside chat.</li>
<li>WooCommerce Cart Actions: Add to cart from chat, refresh cart summary, update quantities, or remove items.</li>
<li>WooCommerce Checkout Assist: Detect checkout requirements, collect required fields, show payment methods, and create an order from chat when your store supports it.</li>
<li>Order Lookup: Surface recent order information for logged-in or identified customers.</li>
<li>WooCommerce-Aware Agent Inbox: Support agents can see useful context such as contact info, last seen page, recently viewed pages, cart snapshot, and recent orders.</li>
<li>Personalized User Context: Logged-in WordPress/WooCommerce users can be recognized automatically for more relevant support.</li>
<li>Live Chat Uploads: Visitors can send images, documents, and voice messages during live chat with staff.</li>
<li>Widget Customization: Choose theme mode, colors, avatars, widget position, welcome message, sound, and custom button image.</li>
<li>Admin Workspace: Includes Dashboard, Account, Settings, Data Sources, Inbox, Playground, and Feedback tabs.</li>
<li>CSV Export: Export leads and full conversation history from the Inbox.</li>
<li>Admin Notifications: Unread chat badge and real-time admin notifications help your team respond faster.</li>
</ul>
This makes the plugin suitable for:
<ul>
<li>business websites that want a knowledge-based AI assistant,</li>
<li>WooCommerce stores that need AI sales/support automation,</li>
<li>teams that want AI first, with a real human ready to step in when necessary.</li>
</ul>
<h3>External Services</h3>
This plugin relies on external services to function. 
By using this plugin, you act as the data controller and agree to send data to the following processors/services:
1. NVT SaaS Platform (<code>api.nguyenvanthu.com</code>) 
* Service: Core AI, indexing, orchestration, and plugin account service. 
* Used For: 
 * Verifying the plugin account and loading account/usage information. 
 * Crawling and indexing website content, products, and uploaded files. 
 * Processing AI chat requests and agentic responses. 
 * Handling feedback submissions sent from the plugin. 
 * Storing or processing chat-related session data required for the service. 
* Data Sent: 
 * Public website content and product data selected for indexing. 
 * Chat messages and conversation context. 
 * Site/domain information and license/account status. 
 * Identified user context such as name, email, phone, or user ID when available. 
 * WooCommerce context when needed, such as products, cart snapshot, checkout context, and recent order data. 
 * Optional feedback attachments and system information when the admin submits feedback. 
* Links: <a href="https://nguyenvanthu.com/terms-of-service/" rel="nofollow ugc">Terms of Service</a> | <a href="https://nguyenvanthu.com/privacy-policy/" rel="nofollow ugc">Privacy Policy</a>
2. NVT Live Service (<code>live.nguyenvanthu.com</code>) 
* Service: Real-time WebSocket/Socket.IO messaging service. 
* Used For: Connecting the visitor widget and the agent Inbox for live chat. 
* Data Sent: Session identifiers, live chat messages, and presence/routing data required for real-time communication.
3. PayPal (<code>paypal.com</code>) 
* Service: Payment processing gateway. 
* Used For: Optional plan upgrades from the plugin Account tab. 
* Links: <a href="https://www.paypal.com/us/legalhub/privacy-full" rel="nofollow ugc">PayPal Privacy Policy</a>
4. PayOS (<code>payos.vn</code>) 
* Service: Payment gateway for Vietnam banking QR payments. 
* Used For: Optional plan upgrades for Vietnam-based payments. 
* Links: <a href="https://payos.vn/thoa-thuan-su-dung/" rel="nofollow ugc">PayOS Terms of Service</a>
<h3>3rd Party Resources</h3>
This plugin bundles or uses the following third-party libraries:
<ul>
<li>Socket.IO Client (v4.8.1)
<ul>
<li>Source: https://github.com/socketio/socket.io-client</li>
<li>License: MIT</li>
</ul>
</li>
<li>Chart.js (v4.4.0)
<ul>
<li>Source: https://github.com/chartjs/Chart.js</li>
<li>License: MIT</li>
</ul>
</li>
</ul>

NVT AI Chatbot – RAG & Live Chat

AI-first RAG chatbot with live agent handoff and deep WooCommerce support for products, cart, checkout, and orders.

mobalabkashima

RAG Chat for Amazon Bedrock is a WordPress plugin that seamlessly integrates your WordPress content with Amazon Bedrock Knowledge Bases to create an intelligent RAG (Retrieval-Augmented Generation) chatbot system.
Key Features:
<ul>
<li>Direct Amazon Bedrock Integration: Connects directly to Amazon Bedrock Knowledge Bases without external servers</li>
<li>Automatic Content Synchronization: Automatically syncs WordPress posts and pages to Amazon Bedrock when content is created, updated, or deleted</li>
<li>Real-time Chat Interface: Provides a chat widget for visitors to interact with your content using AI</li>
<li>Comprehensive Admin Interface: Four dedicated admin pages for configuration and content management</li>
<li>AWS Security: Uses AWS Signature Version 4 authentication for secure API communication</li>
<li>Content Management: Track sync status and manage which content is synchronized to Amazon Bedrock</li>
</ul>
How It Works:
<ol>
<li>Configure your AWS credentials and Amazon Bedrock Knowledge Base in the plugin settings</li>
<li>New WordPress posts and pages are automatically synchronized to Amazon Bedrock when saved or deleted</li>
<li>Existing content can be synchronized using the export function in the Content Management page</li>
<li>Visitors can use the chat interface to ask questions about your content</li>
<li>Amazon Bedrock retrieves relevant content and generates intelligent responses using your WordPress data</li>
</ol>
Requirements:
<ul>
<li>AWS account with Amazon Bedrock access</li>
<li>Amazon Bedrock Knowledge Base configured</li>
<li>Valid AWS credentials (Access Key ID and Secret Access Key)</li>
</ul>
Demo:
See the plugin in action: <a href="https://wp-rag-demo.mobalab.net/ask-rag-chat-for-amazon-bedrock/" rel="nofollow ugc">Live Demo</a>
<h3>Troubleshooting</h3>
Plugin Installation & Setup Issues
<h4>Plugin activation fails</h4>
<ol>
<li>Ensure your WordPress version meets the minimum requirement (6.6+)</li>
<li>Check that your PHP version is 7.4 or higher</li>
<li>Verify there are no plugin conflicts by temporarily deactivating other plugins</li>
</ol>
<h4>The chat interface doesn’t appear on my page</h4>
<ol>
<li>Make sure you’ve inserted the shortcode <code>[rag_chat_ab_chat]</code> in the correct location</li>
<li>Check that the plugin is activated</li>
<li>Verify your AWS credentials are correctly configured in the plugin settings</li>
</ol>
Content Synchronization Issues
<h4>Content sync is failing</h4>
<ol>
<li>Verify your AWS credentials have the correct permissions</li>
<li>Check that your Knowledge Base ID and Data Source ID are correct</li>
<li>Ensure your Knowledge Base has a “Custom” type data source</li>
<li>For existing content, use the “Export Posts and Pages to Amazon Bedrock” feature on the Content Management page</li>
</ol>
<h4>Export operation times out</h4>
If you have many posts and pages, “Export Posts and Pages to Amazon Bedrock” may take a long time and cause timeout errors depending on your PHP settings. To reduce the risk of timeout:
<ol>
<li>Export content in smaller batches by selecting specific date ranges</li>
<li>Consider increasing your PHP execution time limit if you have server access</li>
<li>Contact your hosting provider if timeout issues persist</li>
<li>For large-scale exports, we offer a dedicated bulk export PHP script that runs independently on your server and can handle large volumes of content without web browser timeout limitations. Contact us at <a href="https://tally.so/r/3jjoga" rel="nofollow ugc">https://tally.so/r/3jjoga</a> for more information about our bulk export solution</li>
</ol>
AWS Authentication & Permissions
<h4>AWS permission errors</h4>
<ol>
<li>Double-check all required IAM permissions are granted to your user</li>
<li>Verify your AWS Access Key ID and Secret Access Key are correct</li>
<li>Ensure your AWS account has access to Amazon Bedrock in the selected region</li>
<li>Check the AWS CloudTrail logs for detailed error information</li>
</ol>
<h4>AWS Marketplace subscription error</h4>
If you encounter “Model access is denied” with “aws-marketplace:ViewSubscriptions” permission errors:
<ol>
<li>Your IAM user needs the <code>aws-marketplace:ViewSubscriptions</code> permission</li>
<li>Add this permission to your IAM user’s policy</li>
<li>Wait 15 minutes after fixing the permissions before trying again</li>
<li>This is required for AWS Marketplace subscription completion</li>
</ol>
Model Access & Configuration
<h4>Model use case details required</h4>
If you encounter an error like “Model use case details have not been submitted for this account. Fill out the Anthropic use case details form before using the model”:
<ol>
<li>Anthropic requires use case details before using their models</li>
<li>Fill out the Anthropic use case details form in your AWS console</li>
<li>Wait 15 minutes after submitting before trying again</li>
<li>See the AWS documentation: https://docs.aws.amazon.com/bedrock/latest/userguide/model-access.html</li>
</ol>
<h4>Inference profile required error</h4>
If you get an error like “Invocation of model ID anthropic.claude-haiku-4-5-20251001-v1:0 with on-demand throughput isn’t supported. Retry your request with the ID or ARN of an inference profile”:
<ol>
<li>Some models require using inference profiles instead of direct model invocation</li>
<li>Use the ID or ARN of an inference profile that contains the model</li>
<li>Configure inference profiles in your Amazon Bedrock settings</li>
<li>See the AWS documentation: https://docs.aws.amazon.com/bedrock/latest/userguide/cross-region-inference.html</li>
</ol>
<h4>Chat responses are not working</h4>
<ol>
<li>Verify that your AWS region is correct in the plugin settings</li>
<li>Check that your Knowledge Base contains synchronized content</li>
<li>Ensure your IAM user has <code>bedrock:RetrieveAndGenerate</code> permissions</li>
<li>Try asking questions related to your synchronized content</li>
</ol>
<h3>External services</h3>
This plugin uses the following Amazon Bedrock API endpoints provided by Amazon Web Services, Inc.:
<ul>
<li>PUT https://bedrock-agent.{$region}.amazonaws.com/knowledgebases/{$knowledge_base_id}/datasources/{$data_source_id}
<ul>
<li>Called when a post or page is created or updated.</li>
<li>Used to add / update content on Amazon Bedrock.</li>
<li>The whole content (post or page) is sent.</li>
</ul>
</li>
<li>POST https://bedrock-agent.{$region}.amazonaws.com/knowledgebases/{$knowledge_base_id}/datasources/{$data_source_id}/documents/deleteDocuments
<ul>
<li>Called when a post is deleted or moved to trash.</li>
<li>Used to delete content from Amazon Bedrock.</li>
<li>Only the post / page ID is sent.</li>
</ul>
</li>
<li>POST https://bedrock-agent-runtime.{$region}.amazonaws.com/retrieveAndGenerate
<ul>
<li>Called when a visitor asks a question.</li>
<li>Used to generate responses using Amazon Bedrock’s AI.</li>
<li>The query that the user enters is sent.</li>
<li>The response contains a session ID, and it will be send to this endpoint in the subsequent requests along with the query.</li>
</ul>
</li>
</ul>
Please see the following links for more information about Amazon Web Services:
<ul>
<li><a href="https://aws.amazon.com/service-terms/" rel="nofollow ugc">AWS Service Terms</a></li>
<li><a href="https://aws.amazon.com/privacy/" rel="nofollow ugc">AWS Privacy</a></li>
</ul>
<h3>Related Projects</h3>
This plugin is based on WP RAG (https://github.com/mobalab/wp-rag), which provides RAG functionality using external servers and OpenAI’s APIs. RAG Chat for Amazon Bedrock offers a different approach by integrating directly with Amazon Bedrock Knowledge Bases for users who prefer AWS-native solutions.

RAG Chat for Amazon Bedrock

Integrate WordPress content with Amazon Bedrock Knowledge Bases to create an intelligent RAG (Retrieval-Augmented Generation) chatbot system.

NVT AI Chatbot – RAG & Live Chat vs RAG Chat for Amazon Bedrock