Claude Code Guide
Claude Code, Explained — and How to Start.
Claude Code is Anthropic's agentic coding tool: it reads your repo, edits files, runs tests and iterates from your terminal. It comes with paid Claude plans — Pro (~$20/mo) is the entry — under shared 5-hour rolling limits. Sign up through the invite link below and any bonus Anthropic attaches to invitations applies at signup.
Honest scope
This is a guide, not a rebrand — and VUST is not an IDE agent
Claude Code is Anthropic's product; we link you to it because it's genuinely good at long agentic coding tasks, and the invite link is how this page pays for itself — the bonus terms are Anthropic's, not ours, and we quote no amounts we can't verify. What VUST offers nearby is different in shape: @vustbot answers coding questions, reviews diffs and explains errors per-request in Telegram — a complement or a pre-subscription test bench, not a repo-editing agent.
See the difference
The three things people actually ask: plans, limits, first session.
Who this guide is for
- Deciding developers
- Heard the hype, need the plan/limit facts before subscribing
- The plan ladder, the 5-hour window mechanics and a first-session walkthrough — enough to size the subscription to your real day.
- Invite hunters
- Going to subscribe anyway and want the signup bonus applied
- The invite link on this page applies whatever bonus Anthropic currently attaches — terms are theirs, the link costs you nothing.
- Pre-subscription testers
- Want to feel Claude's coding judgment before paying for a plan
- @vustbot serves Claude-class models per-request in Telegram — review a diff or debug an error today, subscribe when the agent workflow earns it.
From zero to a working session
- 01
Pick the plan by hours, not hype
Pro (~$20/mo) covers focused daily sessions; Max 5x/20x exist for all-day agentic work. Usage is shared with chat and Cowork in rolling 5-hour windows plus weekly caps — Anthropic doubled the Code limits in May 2026.
- 02
Install and run
npm install -g @anthropic-ai/claude-code, then run `claude` inside a repository. It reads the repo, proposes edits, runs tests and iterates — with your review at each step.
- 03
Judge it on one real bug
Hand it a small genuine bug end-to-end: read, edit, test, commit. That loop is 80% of the product — better than any benchmark for deciding whether the plan is worth it.
Test the model before the plan
Ask @vustbot a real coding question with Claude-class models — per request, no subscription. Subscribe to Claude Code when the agent loop earns it.
Honest scope — what this page is and isn't
A guide with an invite link, not a rebrand
Claude Code is Anthropic's product. The invite link is how this page pays for itself; the bonus terms are Anthropic's and we quote no amounts we can't verify at the landing.
VUST is not an IDE agent
@vustbot answers coding questions, reviews diffs and explains errors per-request in Telegram. It does not edit your repository — that honest line is why this page links Claude Code instead of pretending otherwise.
Switch-intent lives elsewhere
If the question is 'do I need the subscription at all' rather than 'how do I start' — the /claude-code-alternative page argues that case with per-request economics.
Frequently asked questions
What is Claude Code, in one paragraph?
An agentic coding tool from Anthropic that runs in your terminal (with IDE integrations): it reads your repository, edits files, runs commands and tests, and iterates until the task is done — driven by the same Claude models as the chat, under your review at each step.
Which plan do I need, and what does it cost?
Claude Code is included in paid Claude plans — there's no standalone subscription. Pro (~$20/month, cheaper billed annually) is the entry tier; Max 5x (~$100) and Max 20x (~$200) raise the usage ceilings for heavy daily use. Free-plan accounts get chat only.
What do I get from the invite link on this page?
Signing up through the invite link applies whatever signup bonus Anthropic currently attaches to invitations — the exact benefit is set by Anthropic and shown at signup, and we deliberately don't quote amounts here because they change. The link never costs you anything extra.
How do the usage limits work in practice?
Two meters: a rolling 5-hour window and a weekly cap, shared across Claude chat, Claude Code and Cowork on one plan. In May 2026 Anthropic doubled the 5-hour Code limits and dropped peak-hour reductions for Pro/Max — so the honest sizing question is hours-per-day of agentic work, not lines of code.
Claude Code vs Cursor — which should I pick?
Different shapes: Cursor is an AI-first IDE you live in; Claude Code is an agent you hand tasks to from the terminal or your existing editor. Terminal-first workflows and long autonomous tasks favor Code; if you want the editor itself to be the AI surface, see our Cursor guide.
Can I try the workflow without subscribing at all?
Not Claude Code itself — it needs a paid plan. But if what you want today is Claude answering coding questions, reviewing a diff or explaining an error, @vustbot serves Claude-class models per-request in Telegram with no subscription — a fair way to test the model's coding judgment before committing to a plan.
Ready when you are
Ready to start? The invite link applies the bonus.
Claude Code comes with paid Claude plans from Pro (~$20/mo). Sign up through the invite link and Anthropic's current signup bonus applies automatically — terms theirs, no cost to you.