> ## Documentation Index
> Fetch the complete documentation index at: https://docs.virtualityhub.com/llms.txt
> Use this file to discover all available pages before exploring further.

# AI code generation

> Describe what you need and V-Run's Claude-powered AI assistant generates, edits, and explains production-ready Python workflow code inside your browser.

V-Run includes a conversational AI assistant powered by Claude that helps you write and maintain Python workflows directly in the browser.

## How it works

Open the **Chat** panel in the workflow editor and describe what you need in plain English. The AI has full context of your workflow — main code, tests, requirements, and config — so it can make targeted suggestions.

Examples of what you can ask:

* "Scrape the top 10 articles from this RSS feed and return them as JSON"
* "Add retry logic with exponential backoff to the API call"
* "Write pytest tests for the `parse_response` function"
* "Explain what the `transform_data` function does"
* "Add a new input parameter for filtering by date"

## Features

### Context-aware

The AI reads your current code, tests, requirements, and config YAML before responding. It understands your workflow's structure and makes suggestions that fit.

### Code generation

Describe a task and the AI generates complete, runnable Python code. Generated code follows best practices and handles common edge cases.

### Iterative refinement

Chat is conversational — you can ask follow-up questions, request changes, or refine the generated code across multiple messages.

### Chat sessions

Each workflow maintains its own chat history. You can create multiple sessions to explore different approaches without losing prior conversations.

<Tip>
  The AI works best when you give it specific, concrete instructions. Instead of "make it better", try "add error handling for HTTP 429 rate limit responses with a 5-second retry".
</Tip>