Chat Interface

The chat interface provides a conversational way to interact with AI directly in your notebook environment.

Opening the Chat Panel

Click the chat icon in the notebook sidebar, or use the keyboard shortcut to toggle the chat panel.

Features

Model Selection

Use the dropdown at the top of the chat panel to select your preferred AI model. Available models depend on your configured providers.

Conversation History

Your chat history persists during your session:

• Scroll up to see previous messages
• Context is maintained across messages
• Use /clear to start fresh

Including Code

Select code in your notebook, then ask questions about it in chat. The selected code is automatically included as context.

Replace Selection

When the AI generates code, you can:

1. Click to copy the code
2. Replace your current selection with the generated code
3. Insert at cursor position

Slash Commands

Type these commands in the chat input:

/generate

Create a new notebook from a description:

/generate Create a data analysis notebook that loads a CSV,
cleans the data, and creates visualizations

The AI generates a complete notebook with code cells and markdown documentation.

/learn

Index local files so you can ask questions about them:

/learn ./src

Options:

• /learn -a ./src - Include all files (not just code)
• /learn -c 500 ./src - Custom chunk size
• /learn -d - Delete the current index

Supported file types:

• Python (.py)
• Markdown (.md)
• Jupyter notebooks (.ipynb)
• Text files (.txt)
• PDF documents (.pdf)
• And more…

/ask

Query content you’ve learned:

/ask How does the authentication module work?

This searches your indexed files and provides answers with source references.

/fix

Fix errors in your code:

1. Run a cell that produces an error
2. Select the cell with the error
3. Type /fix in chat

The AI analyzes the error and suggests a fix.

/clear

Reset the conversation:

/clear

Clears all chat history and starts fresh.

/export

Export your chat history:

/export

Downloads the conversation as a file for reference.

Tips

Be Specific

Good:

Explain how the pandas merge function handles
duplicate keys in left joins

Less effective:

Explain pandas

Use Context

Include relevant code or mention specific functions:

Looking at the process_data function above,
how can I optimize it for large datasets?

Iterate

Build on previous responses:

You: How do I read a CSV in pandas?
AI: [provides example]
You: How do I handle missing values in that data?
AI: [provides follow-up]

Keyboard Shortcuts