MxChat Admin Assistant

AI-powered chat interface built directly into your WordPress admin panel. Work smarter with multiple AI models, image generation, image editing, image analysis, and web search—all without leaving your dashboard.

Quick Setup

1. Install the Plugin: Upload and activate MxChat Admin Assistant through your WordPress plugins page
2. Verify Core Plugin: Ensure MxChat core plugin is installed and active
3. Activate Pro License: Your MxChat Pro license must be active
4. Configure API Keys: Add at least one API key in MxChat → Chatbot → Chat Model Selection
5. Access Admin Chat: Navigate to MxChat → Admin AI Chat in your WordPress admin menu
6. Select Your Model: Click the model selector and choose your preferred AI model

Key Features

Multi-Model AI Chat

Access 20+ AI models from 5 major providers all in one interface. Switch between models instantly to compare responses or use specialized models for specific tasks.

Image Generation

Generate images directly in chat using DALL-E 3, GPT Image, xAI Grok-2-Image, or Google Gemini Nano Banana. Toggle image generation mode with Alt+I or click the image icon.

Image Editing (Gemini)

Upload an image and describe changes you want to make. Gemini Nano Banana will intelligently edit your images based on natural language instructions.

Image Analysis

Upload or paste images for analysis using Grok 2 Vision or GPT-4 Vision. Ask questions about images, extract text, or get detailed descriptions. Toggle with Alt+U.

Web Search (Perplexity)

Enable real-time web search for current information, news, or research. Requires the MxChat Perplexity add-on with configured API key. Toggle with Alt+P.

Chat History Management

Automatically saves all conversations. View, search, load, or delete previous chats. Quick access with the History button in the header.

Auto-Save & Session Recovery

Conversations auto-save every 2 seconds. Sessions persist across page refreshes and browser restarts. Never lose your work.

Available AI Models

How to Use

Basic Chat

1. Navigate to MxChat → Admin AI Chat
2. Select your preferred AI model from the dropdown
3. Type your message in the input field
4. Press Enter to send (or Shift+Enter for new line)
5. View the AI’s response with syntax highlighting for code

Image Generation

1. Click the image icon or press Alt+I to enable image generation mode
2. Configure settings: Click Settings → Choose model (DALL-E, GPT Image, xAI, or Gemini)
3. Describe the image you want to create
4. Press Enter to generate
5. Image appears in chat with download option

Image Editing (Gemini Only)

1. Ensure Gemini is selected as your image generation model in Settings
2. Click the camera icon or press Alt+U
3. Upload or paste an image
4. Click “Edit Mode” button that appears
5. Describe the changes you want (e.g., “make the sky more dramatic”)
6. Press Enter to edit the image
7. Continue editing the result with additional instructions

Image Analysis

1. Click the camera icon or press Alt+U
2. Upload an image or paste from clipboard
3. Ask questions about the image or request analysis
4. Choose between Grok 2 Vision or GPT-4 Vision in Settings
5. Image preview appears above input field
6. Press Enter to analyze

Web Search

1. Install and configure MxChat Perplexity add-on
2. Click the Perplexity icon or press Alt+P to enable
3. Ask questions requiring current information
4. Responses include citations and sources
5. Select Perplexity model in Settings for different response styles

Managing Conversations

• New Chat: Click “New” button to start fresh (auto-saves current chat)
• View History: Click “History” to see all saved conversations
• Load Chat: Click any chat in history to reload it
• Delete Chat: Click trash icon on chat in history
• Search: Use search box in history to find specific conversations

Settings Configuration

Image Generation Settings

Click the Settings button to configure image generation options.

DALL-E 3 Options

• Size: Square (1024×1024), Landscape (1792×1024), Portrait (1024×1792)
• Quality: Standard or HD

GPT Image Options

• Size: 1024×1024, 1536×1024, 1024×1536, or Auto
• Quality: Low, Medium, High, or Auto
• Format: PNG, JPEG, or WebP
• Background: Opaque, Transparent (PNG/WebP only), or Auto
• Compression: 0-100% for JPEG/WebP

xAI Grok Options

• Image Count: Generate 1-10 images per request
• Response Format: URL or Base64 JSON

Gemini Nano Banana

• Text-based prompts only (per Google’s API)
• Includes SynthID watermark for authenticity
• Supports both generation and editing

Image Analysis Settings

• Grok 2 Vision: Fast and accurate with strong visual understanding
• GPT-4 Vision: Detailed analysis with excellent instruction following

Perplexity Settings

• Model Selection: Choose response style (fast, balanced, or detailed)
• Response Mode: Conversational or Structured with citations

API Key Configuration

API keys are configured in the main MxChat settings and shared across all MxChat features including Admin Chat.

OpenAI API Key

1. Go to MxChat → Chatbot → Chat Model Selection
2. Select “OpenAI” from the dropdown
3. Enter your OpenAI API key
4. Click Save
5. You can now use GPT models and DALL-E/GPT Image generation

Anthropic API Key (Claude)

1. Go to MxChat → Chatbot → Chat Model Selection
2. Select any “Claude” model
3. Enter your Anthropic API key
4. Click Save
5. All Claude models are now available

Google Gemini API Key

1. Go to MxChat → Chatbot → Chat Model Selection
2. Select any “Gemini” model
3. Enter your Google API key
4. Click Save
5. Enables Gemini chat models and Nano Banana image generation/editing

xAI API Key (Grok)

1. Go to MxChat → Chatbot → Chat Model Selection
2. Select “xAI” model
3. Enter your xAI API key
4. Click Save
5. Enables Grok models, Grok Vision, and Grok image generation

DeepSeek API Key

1. Go to MxChat → Chatbot → Chat Model Selection
2. Select “DeepSeek” model
3. Enter your DeepSeek API key
4. Click Save

Perplexity API Key (Web Search)

1. Install the MxChat Perplexity add-on
2. Go to MxChat → Perplexity
3. Enter your Perplexity API key
4. Click Save
5. Web search toggle becomes active in Admin Chat

Troubleshooting

Frequently Asked Questions

Pro Tips & Best Practices

Choosing the Right Model

• For coding: GPT-5, Grok Code Fast 1, or Claude 4 Opus
• For writing: Claude 3.7 Sonnet or GPT-4
• For speed: GPT-4o Mini, Gemini 2.0 Flash-Lite, or Claude 3 Haiku
• For complex reasoning: Claude 4 Opus or GPT-5
• For cost-efficiency: DeepSeek-V3 or GPT-3.5 Turbo

Keyboard Shortcuts

• Enter – Send message
• Shift + Enter – New line in message
• Alt + I – Toggle image generation
• Alt + U – Toggle image upload/analysis
• Alt + P – Toggle Perplexity web search

Image Generation Tips

• Be specific and descriptive in your prompts
• Include style, mood, lighting, and composition details
• Use DALL-E 3 for photorealistic images
• Use GPT Image for detailed control over format and quality
• Use xAI Grok for generating multiple variations (1-10 images)
• Use Gemini for images that need text rendering

Image Editing Best Practices

• Start with clear, well-lit images
• Make one change at a time for better results
• Be specific: “make the sky sunset colors” vs “make it better”
• You can edit the same image multiple times iteratively
• Keep images under 10MB for faster processing

Organizing Conversations

• Start a new chat for different topics or projects
• Use descriptive first messages – they become the chat title
• Search history by keywords to quickly find past conversations
• Delete old test chats to keep history organized

Performance Optimization

• Keep conversations focused to avoid context length issues
• Start fresh chats for unrelated topics
• Use faster models for simple queries
• Compress images before uploading for analysis
• Close unused tabs to free up browser memory

Technical Details

System Architecture

• Frontend: jQuery-based interface with modular JavaScript
• Backend: PHP with WordPress AJAX handlers
• Storage: WordPress user meta for conversations, transients for cache
• Images: Server-side file storage in uploads directory
• Session: 24-hour transient cache with localStorage backup

API Timeouts

• Text responses: 300 seconds (5 minutes)
• xAI Grok: 600 seconds (10 minutes)
• Image generation: 120 seconds (2 minutes)
• Image analysis: 60 seconds (1 minute)
• Perplexity search: 180 seconds (3 minutes)

Context Management

• Last 10 messages sent to AI (5 user + 5 assistant)
• Full history stored locally for chat history feature
• Context window prevents token limit errors
• Images not included in context after initial analysis

Security Features

• WordPress nonce verification on all AJAX requests
• Capability checks – requires manage_options
• Input sanitization and output escaping
• User messages stored as plain text, HTML stripped
• Per-user conversation isolation

File Management

• Images saved with unique filenames to prevent conflicts
• Automatic cleanup of images older than 30 days
• .htaccess protection on uploads directory
• Database metadata tracking for all stored images

Browser Compatibility

• Chrome/Edge 90+
• Firefox 88+
• Safari 14+
• Opera 76+

Recent Updates

Need Help?

If you’re experiencing issues not covered in this documentation:

• Visit MxChat.ai for support resources
• Watch tutorials on the MxChat YouTube channel
• Check the browser console (F12) for JavaScript errors
• Verify all plugin requirements are met
• Contact MxChat support with your license key and detailed error description

Useful Links

• OpenAI API Documentation
• Anthropic Claude API Docs
• Google Gemini API Docs
• xAI Grok API Documentation
• DeepSeek API Documentation
• Perplexity API Documentation