AI Coding

Claude Code

An agentic coding AI in your terminal that plans, implements, and fixes autonomously. Built for developers.

Visit official site →
DifficultyIntermediate
PricingIncluded with Claude Pro $20 / Max from $100 / Team $25 per seat
Free planPaid only

What is Claude Code?

The same engine runs in five places - terminal, VS Code, JetBrains, a desktop app, and the browser - and your CLAUDE.md, settings, and MCP servers follow you across all of them. That portability is the real draw: start a task on your laptop, check it from your phone. It suits engineers who already read diffs critically, because it edits files and runs commands on its own. Non-programmers will struggle to review its output. Start with one contained chore, like writing tests for an untested module.

Three things Claude Code does best

  1. One session, five different surfaces

    Terminal CLI, VS Code, JetBrains, desktop app, and claude.ai/code share one engine. Run claude --teleport to pull a web task into your terminal, or /desktop to review diffs visually. Your CLAUDE.md, settings, and MCP servers apply on every surface, so setup carries over.

  2. Delegates work to parallel subagents

    A lead agent can spawn subagents that handle different parts of a task at once, then merge results. Background agents run several full sessions you watch from one screen, and Routines run scheduled jobs in the cloud even when your machine is off.

  3. Fits into scripts and CI

    It follows Unix conventions: pipe logs in, use claude -p for one-shot prompts, chain it with git. GitHub Actions and GitLab CI/CD integrations automate PR review and issue triage, and the Agent SDK lets you build custom agents on the same tooling.

Claude Code is paid only — check the current plans before you commit.

Visit official site →

Recommended for these roles

Skills you can build with Claude Code