How to Use Claude: A Practical Getting Started Guide

Claude is Anthropic's AI model, and there are three main ways to use it: through the Claude.ai web and mobile app for chatting and everyday tasks, through Anthropic's API for building it into your own software, or through a proxy service like SubToAPI if you already have a Claude subscription and want API access without a separate developer account.

Which path makes sense depends on what you're trying to do. If you just want a smart assistant for writing, research, or brainstorming, the chat app is enough. If you're a developer who wants to call Claude from code — to build a chatbot, automate a workflow, or add AI features to a product — you need API access, and that's where the setup gets more involved. This guide walks through both.

Using Claude in the Web or Mobile App

This is the simplest starting point and requires no coding:

1. Go to claude.ai and create an account (or sign in with Google).
2. Pick a plan — there's a free tier with usage limits, and paid plans (Pro, Team) that raise those limits and unlock features like larger file uploads and Projects.
3. Start a conversation. You can paste text, upload documents and images, and ask Claude to summarize, edit, analyze, or generate content.
4. Use Projects to group related conversations and give Claude persistent context (like a style guide or a codebase snippet) so you don't have to repeat it every time.
5. Use the Artifacts panel when Claude generates code, documents, or diagrams — it renders them in a separate pane you can iterate on without losing the chat thread.

This mode is ideal for one-off tasks: drafting emails, reviewing contracts, debugging a script, explaining a concept. It's not built for automation — there's no way to trigger it from your own app or run it on a schedule.

Using Claude Through the API

If you want Claude inside your own product — a support bot, a content pipeline, an internal tool — you need programmatic access. The standard route is Anthropic's API:

1. Create a developer account with Anthropic and generate an API key.
2. Install an SDK or just use HTTP requests.
3. Send a request with a model name, a system prompt (optional), and a list of messages.
4. Parse the response and handle streaming, rate limits, and errors in your code.

A basic request looks like this:

curl https://api.anthropic.com/v1/messages \
  -H "x-api-key: $ANTHROPIC_API_KEY" \
  -H "anthropic-version: 2023-06-01" \
  -H "content-type: application/json" \
  -d '{
    "model": "claude-sonnet-4",
    "max_tokens": 1024,
    "messages": [
      {"role": "user", "content": "Summarize this in two sentences: ..."}
    ]
  }'

This works well, but it means setting up billing separately from any Claude.ai subscription you already pay for, managing your own key rotation, and building your own logging for usage and cost tracking.

Using Claude Through SubToAPI

If you already pay for Claude (Pro, Team, etc.) and just want a clean way to call it from code without duplicating your subscription setup, SubToAPI turns your existing access into a standard HTTPS API. You get an application key (sub_live_...), a dashboard with usage metadata, and support for streaming and tool use — without managing a separate Anthropic developer account for basic integration work.

Getting started:

1. Sign up at /signup and connect your Claude access.
2. Generate a sub_live_... key from the dashboard.
3. Call the API the way you'd call any HTTPS service:

const res = await fetch("https://api.subtoapi.app/v1/messages", {
  method: "POST",
  headers: {
    "Authorization": `Bearer ${process.env.SUBTOAPI_KEY}`,
    "Content-Type": "application/json"
  },
  body: JSON.stringify({
    model: "claude-sonnet-4",
    max_tokens: 1024,
    messages: [
      { role: "user", content: "Draft a one-paragraph product update email." }
    ]
  })
});
const data = await res.json();
console.log(data.content);

The request and response shape follows Anthropic's Messages API conventions, so if you've used Claude's API before, nothing here is unfamiliar — see /docs/messages for the full reference. Streaming responses token-by-token are covered in /docs/streaming, and if you want Claude to call functions in your app (lookups, calculations, external actions), that's in /docs/tools.

For teams, SubToAPI adds seats so multiple developers can share billing and usage visibility under one account instead of juggling individual keys — useful once more than one person is building against Claude. Plans start at Solo (€9), Team (€19/seat), and Scale (€49/seat), all with a free trial — see /pricing for details.

Practical Tips for Using Claude Well

• Be specific in your prompts. Vague requests get vague answers. State the format, length, and audience you want.
• Use system prompts to set persistent behavior — tone, constraints, output format — instead of repeating instructions in every message.
• Break large tasks into steps. For long documents or multi-part analysis, ask Claude to outline first, then fill in sections.
• Check token limits before sending large files. Very long documents may need to be chunked or summarized first.
• Log your requests and responses if you're building a product — it makes debugging model behavior much easier than replaying conversations manually.
• Start with the fastest/cheapest model tier for prototyping, then move to a more capable model once your prompts are stable.

If you're just starting out, the fastest way to get a feel for Claude is the chat app — no setup, immediate feedback. Once you know what you want to build, move to the API. And if you're already paying for Claude and want to skip a second developer account, check the /docs/quickstart guide for getting a SubToAPI key working in minutes.

questions

Do I need to code to use Claude? No. Claude.ai's web and mobile apps let you chat, upload files, and use Projects without writing any code. Coding is only needed if you want to call Claude programmatically from your own application.

What's the difference between using Claude in the app versus the API? The app is for interactive, one-off conversations through a browser or phone. The API lets you send requests from your own code, automate workflows, and integrate Claude's responses into a product — but it requires setup, a key, and request handling.

Can I use my existing Claude subscription for API access? Not directly through Anthropic — API access there requires a separate developer account and billing. SubToAPI bridges that gap by turning your existing Claude access into an HTTPS API key you can call from code; see /docs/quickstart to get started.