Pablo Chatbot User Guide

Pablo Chatbot is an intelligent conversational AI assistant that helps you with a wide range of tasks, from answering questions to creating content, analyzing documents, and solving complex problems. With support for multiple AI models and advanced features like file attachments and conversation management, Pablo is your versatile AI companion.

Table of Contents

1. Overview

2. Interface Layout

3. Starting Conversations

4. Messaging and Interaction

5. File Attachments and Context

6. AI Model Selection

7. Conversation Management

8. Operating Modes

9. Advanced Features

10. Tips and Best Practices

Overview

Pablo Chatbot provides an intuitive interface for interacting with advanced AI models. Whether you need help with writing, coding, research, analysis, or creative tasks, Pablo adapts to your needs and provides intelligent assistance.

Key Features

• Multiple AI Models: Choose from various models including Gemini Pro, Claude, and others

• Conversation History: All your chats are saved and easily accessible

• File Attachments: Upload documents, images, and other files for analysis

• Context Awareness: Pablo remembers conversation context for coherent interactions

• Rich Formatting: Messages support markdown and rich text formatting

• Three Operating Modes: Plan, Act, and Ask modes for different workflows

• Export Capabilities: Save conversations in various formats (MDX, PDF, etc.)

• Search and Organization: Find and organize conversations efficiently

• Multi-Panel Interface: Efficient layout for managing multiple conversations

Interface Layout

Pablo Chatbot features a three-panel interface designed for productive AI interactions:

Left Sidebar - Quick Navigation

The left sidebar provides access to key features:

• Conversations Icon: View all your chat history (primary navigation)

• Tags Icon: Organize conversations with tags

• Globe Icon: Access web-related features

• Calendar Icon: View conversations by date

• Settings Icon: Configure preferences and options

• Additional Tools: Quick access to various features

Center Panel - Conversation List

The conversation list panel shows your chat history:

• Search Bar: Quickly find conversations by content or title

• Conversation Cards: Each showing:

  • Conversation title/preview

  • Timestamp (e.g., “a day ago”, “2 days ago”)

  • Preview of last message

  • Status indicators

• Sorting Options: Organize by date, relevance, or custom order

• Delete Option: Remove conversations with trash icon

• New Conversation: Start fresh chats

Example Conversations:

• “HeadGym Pablo User Guide”

• “Copilot Note”

• “Polymarket Arbitrage Scanner”

• “create a pablo_chatbot_user_guide.mdx”

Right Panel - Chat Interface

The main chat area where conversations happen:

• Status Bar: Shows save status (“Saved”), format (MDX), and notifications

• Model Selector: Choose AI model (e.g., “Gemini Pro”)

• Mode Indicator: Displays current operating mode (Plan/Act/Ask)

• Message Display:

  • User messages (aligned right)

  • Pablo responses (aligned left)

  • Timestamps

  • Formatting and code blocks

• Message Input Area:

  • Text input field with “Message Pablo…” placeholder

  • Attachment button (paperclip icon)

  • Send button

  • Character/token counter

Starting Conversations

Creating a New Conversation

1. Click the “New Conversation” button or press the keyboard shortcut

2. Type your first message in the input field

3. Press Enter or click Send

4. Pablo responds and the conversation begins

5. Conversation is automatically saved with a generated title

Opening Existing Conversations

• Click any conversation in the center panel to open it

• Use the search bar to find specific conversations

• Recently active conversations appear at the top

• Conversations show preview of the last message for quick reference

Conversation Titles

• Conversations are automatically titled based on initial content

• Edit titles by clicking on them in the conversation list

• Use descriptive titles for easy retrieval later

• Example from screenshot: “create a pablo_chatbot_user_guide.mdx”

Messaging and Interaction

Composing Messages

Basic Messaging:

• Type in the input field at the bottom of the chat

• Press Enter to send (or Shift+Enter for new line)

• Messages appear immediately in the chat

• Pablo’s response follows shortly after

Message Formatting:

Use markdown syntax for rich formatting:

• **bold**

  for bold text

• *italic*

  for italic text

• `code`

  for inline code

• ```language

  for code blocks

• # Heading

  for headings

• - item

  for bullet lists

• > quote

  for blockquotes

Multi-line Messages:

• Shift + Enter: Create new line without sending

• Enter: Send message

• Format longer messages with proper structure for clarity

Understanding Pablo’s Responses

Response Format:

• Clear, structured answers

• Code blocks with syntax highlighting

• Bullet points and numbered lists

• Tables and formatted data

• Links and references when relevant

Response Types:

• Direct answers to questions

• Step-by-step instructions

• Code examples and explanations

• Analysis and recommendations

• Creative content generation

• Problem-solving approaches

Interactive Features

Follow-up Questions:

• Pablo maintains conversation context

• Ask clarifying questions

• Build on previous responses

• Iterate and refine answers

Regeneration:

• Request different responses

• Ask for alternative approaches

• Refine and improve outputs

File Attachments and Context

Attaching Files

How to Attach:

1. Click the paperclip icon in the message input area

2. Select files from your computer

3. Or drag and drop files directly into the chat

4. Files appear as attachments in your message

Supported File Types:

• Documents: PDF, DOCX, TXT, MD, MDX

• Images: JPG, PNG, GIF, SVG, WebP

• Code Files: JS, PY, TS, JSON, HTML, CSS, etc.

• Data Files: CSV, XLSX, JSON, XML

• Archives: ZIP, TAR, GZ

• Other: Various text-based formats

Working with Attachments

Document Analysis:

User: [Attaches PDF] "Summarize this document"
Pablo: "This document covers... The main points are:
1. ...
2. ...
3. ..."

Image Understanding:

User: [Attaches screenshot] "What does this interface show?"
Pablo: "This appears to be a chatbot interface with three panels...
The key elements are..."

Code Review:

User: [Attaches code file] "Review this code for issues"
Pablo: "I've analyzed the code. Here are the findings:
- Security: ...
- Performance: ...
- Best practices: ..."

Context Files

As shown in the screenshot, Pablo can access context files:

• HeadGym_Pablo_User_Guide.md

• Pablo_Browser_User_Guide.md

• Pablo_Notes_User_Guide.md

Context Awareness:

• Pablo knows about uploaded and context files

• References file content in responses

• Maintains context across the conversation

• Combines information from multiple sources

• Provides comprehensive answers based on all available context

AI Model Selection

Available Models

The screenshot shows “Gemini Pro” as the current model. Pablo typically supports:

• Gemini Pro: Google’s advanced AI model for general tasks

• Claude Sonnet: Anthropic’s balanced model for writing and analysis

• Claude Opus: Highest capability model for complex tasks

• GPT-4: OpenAI’s advanced model for creative and analytical work

• Other Models: Various specialized models for specific use cases

Selecting a Model

1. Click the model selector dropdown (showing “Gemini Pro”)

2. Browse available models

3. Select the model that fits your needs

4. Model applies to the current conversation

5. You can switch models mid-conversation if needed

Model Characteristics

Gemini Pro:

• Strong general-purpose capabilities

• Good for analysis and reasoning

• Multimodal (text and images)

• Fast response times

• Excellent for research and information

Claude Models:

• Excellent for writing and content creation

• Strong coding capabilities

• Thoughtful, nuanced responses

• Good context understanding

• Great for analysis and planning

GPT Models:

• Versatile and creative

• Strong coding assistance

• Broad knowledge base

• Plugin ecosystem

• Good for diverse tasks

Choosing the Right Model

For Writing:

• Claude Sonnet or Opus for nuanced, high-quality content

• GPT-4 for creative writing and storytelling

For Coding:

• Claude Sonnet for complex code and architecture

• Gemini Pro for quick solutions and debugging

For Analysis:

• Claude Opus for deep, comprehensive analysis

• Gemini Pro for data interpretation and research

For Speed:

• Gemini Pro for quick responses

• Smaller models for simple tasks

Conversation Management

Organizing Conversations

Search Function:

• Search bar at top of conversation list

• Search by content, title, or keywords

• Filter by date or tags

• Quick access to any conversation in your history

Sorting and Filtering:

• Most recent conversations appear first

• Filter by date range

• Group by tags or topics

• Custom organization options

Deleting Conversations

• Click trash icon on conversation card

• Confirm deletion when prompted

• Permanently removes conversation

• Cannot be undone - use with caution

Conversation Status

The status bar displays:

• Saved: Conversation is automatically saved

• Format: Current export format (e.g., MDX)

• Notifications: Alerts and updates

• Sync Status: Cloud synchronization indicator

Exporting Conversations

Export Formats:

• MDX: Markdown with JSX (shown in screenshot)

• PDF: Formatted document for sharing

• Markdown: Plain markdown for editing

• HTML: Web format for publishing

• JSON: Structured data for processing

Export Process:

1. Open conversation to export

2. Click export option in menu

3. Select desired format (MDX, PDF, etc.)

4. Choose save location

5. File is generated and saved to your device

Operating Modes

Pablo Chatbot supports three distinct operating modes, each designed for different types of interactions:

Plan Mode

Purpose: Strategic planning and brainstorming before implementation

Use When:

• Planning a project or complex task

• Brainstorming ideas and approaches

• Architecting solutions before coding

• Discussing strategies and methodologies

Features:

• Conversational planning and discussion

• No direct actions or file modifications

• Strategy discussion and refinement

• Outline and roadmap creation

• Collaborative thinking

Example Interaction:

User: "I need to build a web application for task management"
Pablo: "Let's plan this together. What are the main features you need? 
We should consider:
1. User authentication and authorization
2. Database structure for tasks and users
3. Frontend framework (React, Vue, or Angular?)
4. Backend API design
5. Deployment strategy

What's most important for your use case? Who are your target users?"

When to Use Plan Mode:

• Starting new projects

• Complex problem-solving

• Architectural decisions

• Before switching to Act Mode for implementation

• When you want to discuss and refine ideas

Act Mode

Purpose: Active execution and implementation of solutions

Use When:

• Creating or modifying files
• Executing commands and operations
• Implementing solutions from plans
• Making changes to systems and code

Features:

• Direct file creation and editing
• Command execution capabilities
• Tool usage for various operations
• Implementation of discussed plans
• Real-time action and feedback

Example Interaction:

User: "Create the user guide based on this screenshot"
Pablo: [Analyzes screenshot, creates file structure, writes comprehensive content]
Result: pablo_chatbot_user_guide.mdx created with complete documentation

When to Use Act Mode:

• After planning is complete
• When you need actual changes implemented
• For file operations and modifications
• To execute discussed solutions
• When ready to build or create

Ask Mode

Purpose: Information gathering, learning, and general conversation

Use When:

• Asking questions about any topic
• Learning new concepts or technologies
• Seeking explanations and clarifications
• General conversation and exploration
• Research and information retrieval

Features:

• Question answering on any topic
• Explanations and tutorials
• Information retrieval from knowledge base
• No file modifications or actions
• Conversational and educational

Example Interaction:

User: "How does OAuth 2.0 work?"
Pablo: "OAuth 2.0 is an authorization framework that enables applications 
to obtain limited access to user accounts on an HTTP service. Here's how it works:

1. The user initiates the process by clicking 'Login with...'
2. The application redirects to the authorization server
3. User authenticates and grants permissions
4. Authorization server issues an access token
5. Application uses token to access protected resources

Key components include:
- Resource Owner (user)
- Client (application)
- Authorization Server
- Resource Server"

When to Use Ask Mode:

• Learning and research
• Quick questions and clarifications
• Exploring ideas without implementation
• When you don’t need actions or changes
• General knowledge queries

Switching Between Modes

Mode Selection:

• Mode selector visible in the interface
• Click to switch between modes
• Current mode indicated clearly in status bar
• Interface adapts to selected mode

Recommended Workflow:

1. Start in ASK MODE: Learn about the topic or technology
2. Switch to PLAN MODE: Discuss and plan the approach
3. Move to ACT MODE: Implement the solution
4. Return to ASK MODE: Learn from results and iterate

Mode Indicators:

• Current mode shown in chat interface
• Different capabilities available per mode
• Clear transitions between modes
• Context maintained across mode switches

Advanced Features

Multi-File Operations

When working on complex tasks, Pablo can:

• Create multiple files in sequence
• Modify several files simultaneously
• Maintain consistency across files
• Handle project-wide changes

Example:

User: "Create a React component with tests and documentation"
Pablo: Creates:
- Component.tsx
- Component.test.tsx
- Component.md
- index.ts

Code Generation and Analysis

Code Creation:

• Generate complete applications
• Create components and modules
• Write tests and documentation
• Follow best practices and patterns

Code Review:

• Analyze existing code
• Identify bugs and issues
• Suggest improvements
• Explain complex logic

Refactoring:

• Modernize legacy code
• Improve performance
• Enhance readability
• Apply design patterns

Document Processing

Analysis:

• Summarize long documents
• Extract key information
• Compare multiple documents
• Identify patterns and insights

Generation:

• Create comprehensive documentation
• Write user guides and tutorials
• Generate reports and summaries
• Produce formatted content

Research and Information

Web Research:

• Find current information
• Synthesize multiple sources
• Provide cited references
• Stay up-to-date with latest developments

Data Analysis:

• Process CSV and spreadsheet data
• Generate visualizations
• Identify trends and patterns
• Provide statistical insights

Collaboration Features

Sharing:

• Export conversations for sharing
• Generate shareable links
• Create documentation from chats
• Collaborate asynchronously

Version Control:

• Track conversation history
• Review previous iterations
• Compare different approaches
• Maintain audit trail

Tips and Best Practices

Effective Communication

Be Specific:

❌ "Help me with my code"
✅ "Review this React component for performance issues and suggest optimizations"

Provide Context:

❌ "Fix this error"
✅ "I'm getting a TypeError in my Node.js app when trying to access user.email. 
Here's the relevant code: [code snippet]"

Break Down Complex Tasks:

Instead of: "Build me a complete e-commerce platform"
Try: 
1. "Let's plan the architecture for an e-commerce platform"
2. "Create the user authentication system"
3. "Implement the product catalog"
4. "Build the shopping cart functionality"

Using Attachments Effectively

Best Practices:

• Attach relevant files only
• Use clear file names
• Provide context about what you need
• Specify what aspects to focus on

Examples:

Good: [Attaches screenshot] "Analyze the layout and suggest UI improvements"
Good: [Attaches code] "Review the authentication logic in lines 45-78"
Good: [Attaches PDF] "Extract the key findings from the executive summary"

Optimizing for Different Tasks

For Writing:

• Specify tone and style
• Provide target audience
• Mention length requirements
• Share examples if available

For Coding:

• Specify programming language
• Mention framework/library versions
• Share relevant error messages
• Provide context about the project

For Analysis:

• Define what you want to know
• Specify depth of analysis
• Mention any constraints
• Ask for specific formats

Managing Long Conversations

Stay Organized:

• Use clear subject changes
• Reference previous messages when needed
• Create new conversations for new topics
• Use tags to categorize conversations

Maintain Context:

• Remind Pablo of key points if needed
• Reference earlier decisions
• Build on previous responses
• Keep related discussions together

Leveraging Operating Modes

Plan Mode Strategy:

• Start complex projects here
• Discuss trade-offs and options
• Get feedback before implementation
• Create detailed roadmaps

Act Mode Strategy:

• Have a clear plan first
• Review changes before approving
• Test implementations thoroughly
• Iterate based on results

Ask Mode Strategy:

• Learn before planning
• Clarify concepts and approaches
• Research best practices
• Explore possibilities

Privacy and Security

Best Practices:

• Don’t share sensitive credentials
• Avoid personal identifying information
• Review generated code for security
• Be cautious with production data

Data Handling:

• Conversations are saved automatically
• Export important conversations
• Delete sensitive conversations when done
• Use appropriate access controls

Performance Tips

For Faster Responses:

• Ask focused questions
• Use appropriate model for task
• Avoid unnecessary attachments
• Keep conversations on topic

For Better Results:

• Provide complete context
• Be specific about requirements
• Iterate and refine responses
• Give feedback on outputs

Common Use Cases

Content Creation:

• Blog posts and articles
• Documentation and guides
• Marketing copy
• Social media content

Software Development:

• Code generation
• Bug fixing
• Code review
• Architecture design

Learning and Research:

• Concept explanations
• Tutorial creation
• Research summaries
• Skill development

Business Tasks:

• Report generation
• Data analysis
• Process documentation
• Decision support

Troubleshooting

If Pablo Doesn’t Understand:

• Rephrase your question
• Provide more context
• Break into smaller parts
• Share examples

If Results Aren’t Satisfactory:

• Ask for alternatives
• Provide more specific requirements
• Try a different model
• Iterate with feedback

If Experiencing Technical Issues:

• Refresh the page
• Check your connection
• Try a different browser
• Clear cache if needed

Keyboard Shortcuts

Common shortcuts for efficient use:

• Enter: Send message
• Shift + Enter: New line in message
• Ctrl/Cmd + K: New conversation
• Ctrl/Cmd + F: Search conversations
• Ctrl/Cmd + /: Show shortcuts

Getting Help

In-App Resources:

• Access help from settings
• View keyboard shortcuts
• Check documentation
• Contact support if needed

Learning Resources:

• User guides (like this one)
• Example conversations
• Video tutorials
• Community forums

Conclusion

Pablo Chatbot is a powerful AI assistant designed to help you accomplish a wide variety of tasks efficiently. By understanding the interface, leveraging the three operating modes, and following best practices, you can maximize your productivity and get the most out of your interactions with Pablo.

Key Takeaways:

• Use the three-panel interface for efficient navigation
• Choose the right AI model for your task
• Leverage operating modes (Plan, Act, Ask) appropriately
• Attach files and provide context for better results
• Organize conversations with search and tags
• Export important conversations for future reference

Next Steps:

• Explore the interface and features
• Try different AI models
• Experiment with file attachments
• Practice using different operating modes
• Build your workflow with Pablo

Happy chatting with Pablo! 🤖✨