ChatProjects Documentation

ChatProjects Pro is a WordPress plugin that transforms your website into a powerful AI workspace. Unlike SaaS chat tools that require monthly subscriptions and store your data on external servers, ChatProjects lets you:

• Use your own API keys - Pay only for what you use, directly to AI providers
• Keep data on your server - All conversations stored in your WordPress database
• Chat with 5+ AI providers - Switch between GPT-5, Claude, Gemini, DeepSeek, and 100+ models via OpenRouter
• Create knowledge-based projects - Upload documents and chat with their contents using Vector Stores and the Responses API
• White-label for clients - Full branding customization for agencies PRO

Who is ChatProjects For?

Why ChatProjects vs. ChatGPT or Claude Directly?

Server Requirements

Required PHP Extensions

• OpenSSL - For API key encryption (AES-256)
• cURL - For API communication with AI providers

Memory & Upload Settings

Browser Support

Installing the Plugin

Method 1: WordPress Plugin Directory (Free Version)

1. Go to Plugins > Add New in your WordPress admin
2. Search for "ChatProjects"
3. Click Install Now
4. Click Activate

Method 2: Manual Upload (Pro Version)

1. Download the Pro plugin ZIP file from your account
2. Go to Plugins > Add New > Upload Plugin
3. Choose the ZIP file and click Install Now
4. Click Activate

Method 3: FTP Upload

1. Download and extract the plugin ZIP file
2. Upload the chatprojects-pro folder to /wp-content/plugins/
3. Go to Plugins in WordPress admin
4. Click Activate next to ChatProjects Pro

Initial Configuration Checklist

1. Navigate to ChatProjects > Settings in the WordPress admin menu
2. Add at least one API key - OpenAI is required for project features
3. Configure default provider - Select your preferred AI provider
4. Set default model - Choose the model for new conversations
5. Configure file upload limit - Adjust if needed (default: 50MB)
6. Create a page - Add the [chatprojects_main] shortcode to any page
7. Test the interface - Visit your page and send a test message

Upgrading from Free to Pro

1. Deactivate the free ChatProjects plugin (do not delete)
2. Install ChatProjects Pro using Method 2 or 3 above
3. Activate ChatProjects Pro
4. Your settings, projects, and chat history are automatically preserved

You need at least one API key to use ChatProjects. OpenAI is required for project features (Vector Stores and file search).

OpenAI API Key

Required for: Projects, Vector Stores, File Search, GPT models

1. Visit platform.openai.com
2. Sign up or log in to your account
3. Go to API Keys in the left sidebar
4. Click Create new secret key
5. Copy the key (starts with sk-)
6. Paste into ChatProjects Settings

Pricing: Pay-per-use based on tokens. See openai.com/pricing

Anthropic API Key

Required for: Claude models

1. Visit console.anthropic.com
2. Sign up or log in
3. Go to Settings > API Keys
4. Create a new key (starts with sk-ant-)
5. Copy and paste into ChatProjects Settings

Google Gemini API Key

Required for: Gemini models, Image Studio PRO

1. Visit ai.google.dev
2. Sign in with your Google account
3. Click Get API Key
4. Create a key in a new or existing project (starts with AIza)
5. Copy and paste into ChatProjects Settings

Chutes API Key

Required for: DeepSeek models

1. Visit chutes.ai
2. Sign up or log in
3. Navigate to API settings
4. Generate your API key
5. Copy and paste into ChatProjects Settings

OpenRouter API Key

Required for: 100+ models from multiple providers

1. Visit openrouter.ai
2. Sign up or log in
3. Go to Keys section
4. Create a new key (starts with sk-or-)
5. Copy and paste into ChatProjects Settings

OpenAI Models

Anthropic Claude Models

Google Gemini Models

Chutes Models

OpenRouter

OpenRouter provides access to 100+ models from multiple providers through a single API, including models from OpenAI, Anthropic, Google, Meta, Mistral, and more.

Chat Modes

Starting a Conversation

1. Select Provider - Use the dropdown to choose your AI provider
2. Select Model - Choose from available models for that provider
3. Type your message - Enter your question or prompt
4. Press Enter or click Send

Messages stream in real-time as the AI generates responses.

Managing Chat History

Auto-Generated Titles

Chat titles are automatically generated from your first message. You can rename them at any time.

Renaming Chats

1. Click the chat name in the sidebar
2. Enter a new name
3. Press Enter to save

Deleting Chats

1. Hover over a chat in the sidebar
2. Click the delete icon (trash)
3. Confirm deletion

Switching Providers Mid-Conversation

You can switch AI providers or models at any point in a conversation. The new provider/model will be used for that message and subsequent ones. Your conversation context is maintained.

Dark/Light Mode

ChatProjects includes full dark mode support:

• Automatic detection - Follows your system preference by default
• Manual toggle - Click the theme toggle in the header
• Persistent preference - Your choice is saved for future sessions

Projects are organized workspaces where you can upload documents and have AI-powered conversations about their contents.

Creating a New Project

1. Navigate to the Projects tab
2. Click New Project
3. Fill in the project details:

   Field	Description	Required
   Title	Project name (e.g., "Product Documentation")	Yes
   Description	What this project contains	No
   Instructions	Custom AI behavior for this project	No

4. Click Create Project

Project Custom Instructions

Custom instructions define how the AI should behave when answering questions about your project documents.

Example: Technical Support

You are a technical support agent for our software product.
Answer questions based on the uploaded documentation.
Be concise and cite specific document sections when possible.
If you don't find the answer in the documents, say so.

Example: Research Assistant

You are a research assistant helping analyze uploaded papers.
Summarize findings, compare methodologies, and identify key conclusions.
Cite specific papers and page numbers when referencing information.

Project Sharing PRO

Free vs Pro Project Limits

RAG (Retrieval-Augmented Generation) allows the AI to search your uploaded documents and provide answers based on their contents.

How Vector Stores Work

When you upload files to a project:

1. Upload - Files are sent to OpenAI's servers
2. Processing - OpenAI extracts and chunks the text content
3. Embedding - Text is converted to vector embeddings
4. Indexing - Vectors are stored in your project's Vector Store
5. Search - When you chat, AI searches for relevant content
6. Answer - AI uses found content to generate contextual answers

Supported File Types

Documents

Data Files

Code Files

Spreadsheets

File Size Limits

Uploading Files

1. Open your project
2. Go to the Files tab
3. Drag and drop files onto the upload area, or click to browse
4. Wait for the upload progress to complete
5. Files are automatically indexed (may take a few moments for large files)

Embed ChatProjects anywhere on your WordPress site using shortcodes.

Main Application Shortcode

[chatprojects_main]

Renders the full ChatProjects application including navigation tabs, sidebar, chat interface, and project management.

Attributes

Examples

[chatprojects_main]
[chatprojects_main default_tab="projects"]
[chatprojects_main height="600px"]

Widget Shortcode (Floating Chat)

[chatprojects_widget]

Renders a floating chat widget that can be positioned anywhere on the page.

Widget Attributes

Widget Examples

[chatprojects_widget]
[chatprojects_widget position="bottom-left" height="600px" width="450px"]
[chatprojects_widget theme="dark" collapsed="true"]
[chatprojects_widget draggable="true" button_icon="robot" button_color="#10b981"]

Build and manage a library of reusable AI prompts for consistent, high-quality interactions.

What's Included

• 27 professionally-crafted default prompts covering common use cases
• 8 categories for organization: General, Assistant, Code Generation, Content Writing, Analysis, Creative, Business, Education

Default Prompt Categories

Features

• Create Custom Prompts - Add your own prompts with title, category, and content
• AI-Powered Enhancement - Click Enhance to optimize your prompt using GPT-5.2
• Template Variables - Use variables like {{word_count}}, {{topic}}, {{audience}}
• Team Sharing - Toggle sharing to make prompts available to other users
• One-Click Insert - Click any prompt to instantly insert it into your chat input

Template Variable Example

Write a {{word_count}} word blog post about {{topic}}.
Target audience: {{audience}}.
Tone: {{tone}}.

Convert audio and video recordings to text using OpenAI's Whisper API, with AI-powered tone rewriting.

Supported Formats

Maximum file size: 25 MB

How to Transcribe

1. Go to the Transcribe tab
2. Click Upload Audio or drag and drop your file
3. Wait for processing (typically a few seconds)
4. View your transcription

Tone Rewriting

Transform your transcription with one-click tone options:

Compare responses from two AI models side-by-side to find the best fit for your needs.

Features

• Any Two Models - Compare across providers (e.g., GPT-5 vs Claude)
• Parallel Responses - Send the same prompt, get both responses simultaneously
• Response Timing - See which model responds faster
• Full History - Complete conversation preserved for both models
• Session Management - Save, browse, and delete comparison sessions

How to Compare

1. Go to the Compare tab
2. Select Model A (provider and model)
3. Select Model B (provider and model)
4. Type your prompt
5. Send to see both responses

Use Cases

• Evaluate which model writes better code
• Compare explanation quality
• Test creative writing styles
• Assess accuracy on factual questions
• Find the best value (quality vs. cost)

Give AI access to current web information for up-to-date answers.

How It Works

• Provider: DuckDuckGo (free, no API key required)
• Results: 1-20 results per search (default: 5)
• Integration: Search results are included in AI context

Auto-Trigger Keywords

The AI automatically searches the web when your message contains:

• "current", "latest", "recent", "today"
• Years (2024, 2025, etc.)
• "news", "weather", "stock price"
• "search for", "look up", "find information"

Configuration

1. Go to ChatProjects > Settings
2. Find the Web Search section
3. Enable/disable web search
4. Adjust number of results (1-20)

Install ChatProjects as a native app on any device for an app-like experience.

Supported Platforms

• iOS - Add to Home Screen from Safari
• Android - Install prompt in Chrome
• Windows - Install from Edge or Chrome
• macOS - Install from Chrome

Features

PWA Customization

Configure your PWA in ChatProjects > Settings > PWA:

Complete customization for agencies and resellers. Remove all ChatProjects branding and replace with your own.

Customization Options

Where Branding Appears

Your custom branding is applied to:

• Admin interface headers
• Frontend chat interface
• Page titles and headings
• PWA manifest and icons
• Loading screens

CSS Variables

ChatProjects generates CSS variables you can use in custom CSS:

:root {
  --cp-primary: #your-primary-color;
  --cp-secondary: #your-secondary-color;
  --cp-accent: #your-accent-color;
  --cp-primary-rgb: r, g, b;  /* For opacity */
}

Create and edit images through natural language conversation.

Multi-Turn Editing (Gemini)

When Google Gemini API is configured, Image Studio provides a conversational editing experience:

1. Describe the image you want to create
2. View the generated result
3. Request changes: "make the sky more dramatic", "add a sunset glow", "remove the person on the left"
4. Continue refining through conversation
5. Download or save to Media Library

Aspect Ratios

OpenAI Fallback

When Gemini is not configured, single-shot generation is available via OpenAI:

• DALL-E 3 - 1024x1024, 1792x1024, 1024x1792 (HD quality)
• GPT Image 1 - 1024x1024, 1024x1536, 1536x1024 (Fast generation)

Quality Options

Auto (recommended), Low (fastest), Medium, High, Standard, HD (highest quality)

Export your conversations in multiple formats for documentation, sharing, or backup.

Export Formats

Features

• Full or Partial Export - Export entire conversation or up to a specific message
• Metadata Included - Provider, model, date, message count
• Sources Preserved - Document references maintained
• Professional Styling - HTML export includes dark/light CSS themes

How to Export

1. Open the conversation you want to export
2. Click the Export button (download icon)
3. Select format (Markdown, JSON, or HTML)
4. Choose full or partial export
5. Download the file

Monitor API usage and track costs across all providers.

Metrics Tracked

Dashboard Views

Personal Dashboard: Each user sees their own usage metrics and costs.

Platform Analytics (Admin): Administrators see aggregate metrics across all users: total usage by model, cost trends over time, top users by usage, model popularity rankings.

Date Ranges

Last 7 days, Last 30 days, Last 90 days, All time

Configurable Pricing

Set custom pricing per model to match your actual costs:

1. Go to ChatProjects > Settings > Pricing
2. Enter input token rate (per 1K tokens)
3. Enter output token rate (per 1K tokens)
4. Costs are calculated automatically

API Key Encryption

All API keys are encrypted before storage using industry-standard encryption:

Your API keys are:

• Never stored in plain text
• Decrypted only during API calls
• Never transmitted to ChatProjects servers
• Never displayed after initial entry

Data Storage

Data Transmission

ChatProjects connects directly to AI providers - there is no middleman server:

• Your WordPress site > AI Provider (direct connection)
• Only the provider you select receives your data
• No data passes through ChatProjects servers

WordPress Security Standards

Frequently Asked Questions

Common Issues

ChatProjects connects to external AI services. Your data is transmitted according to each provider's privacy policy.

Documentation

Support Channels

Additional Links

Version 1.0.0

Initial Release - December 2025

Core Features

• Multi-provider AI chat (OpenAI, Anthropic, Gemini, Chutes, OpenRouter)
• Project management with OpenAI Vector Stores
• File upload and AI-powered document search (RAG)
• Modern chat interface with real-time streaming
• Dark/Light theme support
• AES-256 API key encryption
• Local message storage
• Shortcode embedding

Pro Features

• Prompt Library with 27 default prompts
• Audio transcription with tone rewriting
• Side-by-side model comparison
• DuckDuckGo web search integration
• Progressive Web App (PWA) support
• White-label branding customization
• Image Studio (Gemini multi-turn + OpenAI)
• Chat export (Markdown, JSON, HTML)
• Usage analytics and cost tracking

Supported File Types

PDF, DOC, DOCX, TXT, MD, PPTX, CSV, JSON, XML, HTML, CSS, JS, PY, PHP, Java, C++, XLS, XLSX, ODS