Overview

APL1 is the starting point. You talk to an AI assistant through a browser. It cannot touch your files, your code, or your system. The risk is low - but so is the leverage.

What This Level Is

Zero Setup means exactly that - no software to install, no terminal to open, no API keys to manage. You open a browser tab, type a message, and get a response. This is how most people first interact with an AI assistant.

Tools at This Level

  • ChatGPT - OpenAI's chat interface
  • Claude.ai - Anthropic's conversational assistant
  • Gemini Chat - Google's AI assistant (chat interface)
  • Grok - xAI's AI assistant
  • DeepSeek Chat - DeepSeek's chat interface (same model via Cline = APL3)
  • Minimax Chat - Minimax AI chat interface
  • z.ai - Zhipu AI's chat interface (GLM models)
  • Perplexity - AI-powered search with citations
  • Lovable / Bolt.new - Browser-based app builders (no local setup)

What You Can Do

  • Ask questions and get explanations
  • Generate text, emails, outlines
  • Brainstorm ideas and get feedback
  • Copy-paste code snippets for review
  • Learn concepts through conversation

What You Cannot Do

  • Edit files directly - you copy/paste manually
  • Access your codebase - the assistant has no context about your project
  • Run commands - there is no terminal connection
  • Automate anything - every interaction requires manual effort

APL1 is the safest level because the assistant has no access to anything outside the chat. But there are still considerations:

Risks at APL1

  • Data leakage: What you paste into the chat may be used for training (check provider policies)
  • Over-reliance: Trusting output without verification
  • Stagnation: Staying at APL1 when your needs have outgrown it

When to Move Up

If you are spending more time copy-pasting between your editor and a chat window than actually building, it is time to look at APL2: Light Setup.