​
What is the Chat Assistant?

The Chat Assistant is a lightweight interface for asking questions about your code, getting explanations, and performing simple coding tasks. Unlike generic AI assistants, it understands your codebase through Repo Grokking, allowing it to provide contextually relevant answers.

​
How to Use the Chat Assistant

1

Access the Chat Interface

Open the Zencoder panel in your IDE (typically in the sidebar or through the Zencoder icon).

2

The Chat Assistnt is Great for...

  • Code explanations
What does this function do?

Explain how this API endpoint works"
  • Error understanding
Why am I getting this TypeError?"
  • Simple code generation
Write a function to validate email addresses
  • Best practices
What's the best way to handle this exception?
3

Use Code References

Reference specific files, functions, or classes in your questions for more precise answers:

"How does the authentication work in auth_service.py?"

"What's the purpose of the UserManager class?"

​
Chat Assistant vs. Coding Agent

While both interfaces are powered by the same technology, they serve different purposes:

Chat AssistantCoding Agent
Quick questions and explanationsComplex multi-file operations
Simple code snippetsFull feature implementation
Limited file accessFull repository access
No external toolsWeb search and other tools
Always availableMust be explicitly enabled

​
Technical Details

  • The Chat Assistant uses the same codebase understanding as the Coding Agent but with limited abilities
  • It can read files but has restricted ability to modify them
  • It doesn’t have full access to external tools like web search
  • It’s optimized for quick responses rather than complex problem-solving

​
Enabling the Coding Agent

When you need more powerful capabilities:

  1. Look for the toggle switch in the chat interface (typically labeled “Coding agent”)
  2. Enable the toggle to transform the Chat Assistant into the Coding Agent
  3. The Coding Agent can perform more complex tasks like:
    • Creating and modifying multiple files
    • Implementing entire features
    • Searching the web for documentation
    • Running validation and tests