Claude Code
Claude Code is Anthropic’s agentic coding tool. It reads your codebase, edits files across multiple files, runs terminal commands, and integrates with your development workflow — from Terminal to CI/CD.
Installation
Terminal (Recommended)
# macOS, Linux, WSLcurl -fsSL https://claude.ai/install.sh | bash
# Windowsirm https://claude.ai/install.ps1 | iex
# Homebrewbrew install --cask claude-code
# WinGetwinget install Anthropic.ClaudeCodeVS Code
Install “Claude Code” from the Extensions marketplace (Cmd+Shift+X). Provides inline diffs, @-mentions, plan review.
JetBrains
Install the Claude Code plugin for IntelliJ, PyCharm, WebStorm.
Desktop App
Download from claude.ai — visual diffs, side-by-side sessions, scheduled tasks.
Web
Go to claude.ai/code — no setup, browser-based, works on iOS.
CI/CD
# GitHub Actions- uses: anthropics/claude-code-action@v1 with: prompt: "Review this PR for security issues"Also available for GitLab CI/CD.
Seven Surfaces — One Engine
| Surface | Best For | Key Feature |
|---|---|---|
| Terminal CLI | Full control, scripting, CI/CD | claude -p "prompt" for quick tasks |
| VS Code | IDE integration, inline diffs | @-mentions, plan mode, context sharing |
| JetBrains | IntelliJ, PyCharm, WebStorm | Interactive diff viewing |
| Desktop App | Visual workflow, multi-session | Side-by-side sessions, recurring tasks |
| Web | No setup, browser/phone | Remote execution, long-running cloud tasks |
| iOS | On-the-go coding | Dispatch tasks, review results |
| CI/CD | Automated PR review, issue triage | GitHub Actions, GitLab CI/CD |
Core Capabilities
Read, Write, Execute
# Explore and understandclaude "explain the authentication flow in this codebase"
# Build featuresclaude "add a rate-limiting middleware to the API"
# Fix bugsclaude "fix the race condition in the payment processing module"
# Write testsclaude "write tests for the auth module, run them, and fix any failures"
# Git operationsclaude "commit my changes with a descriptive message and open a PR"MCP Integration
Connect Claude Code to external tools and data:
# Configure MCP servers in .claude/mcp.json{ "mcpServers": { "filesystem": { "command": "npx", "args": ["-y", "@modelcontextprotocol/server-filesystem"] }, "github": { "command": "npx", "args": ["-y", "@modelcontextprotocol/server-github"] } }}See MCP — Model Context Protocol for a complete guide.
Agent Skills (SKILL.md)
Package repeatable workflows your team can share:
# Project skills in .claude/skills/.claude/skills/ generate-migration/ SKILL.md # Instructions for the skill review-pr/ SKILL.md deploy-staging/ SKILL.mdSee Agent Skills for details.
CLAUDE.md — Persistent Instructions
Add CLAUDE.md to your project root:
## Code Style- Use TypeScript strict mode- Prefer functional patterns over classes- Tests go in __tests__/ colocated with source
## Architecture- API routes in src/api/- Business logic in src/services/Claude Code reads this at the start of every session. It also builds auto-memory — learning build commands, debugging insights, and project conventions as it works.
Hooks
Run shell commands before or after Claude Code actions:
{ "hooks": { "PostToolUse": [ { "matcher": "Edit|Write", "hooks": [{"type": "command", "command": "prettier --write ${CLAUDE_TOOL_INPUT_FILE_PATH}"}] } ], "PreToolUse": [ { "matcher": "Bash(git commit*)", "hooks": [{"type": "command", "command": "npm run lint"}] } ] }}Sub-Agents — Parallel Task Execution
Spawn multiple Claude Code agents working simultaneously:
claude "audit all services for security issues, checking each in parallel"A lead agent coordinates, assigns subtasks to sub-agents, and merges results.
Routines & Scheduling
Run Claude on a schedule — even when your computer is off:
- Routines: Runs on Anthropic infrastructure. Trigger by schedule, API call, or GitHub events.
- Desktop scheduled tasks: Run on your machine with local file access.
/loop: Repeat a prompt within a CLI session for quick polling.
Remote Control & Channels
- Remote Control: Continue sessions from your phone or any browser
- Channels: Push events from Telegram, Discord, iMessage, or custom webhooks into sessions
- Dispatch: Send tasks from your phone, execute on your desktop
Claude Code vs Cursor vs Copilot
| Feature | Claude Code | Cursor | GitHub Copilot |
|---|---|---|---|
| Primary interface | Terminal CLI | IDE fork (VSCode-based) | IDE extension |
| Model | Claude Opus 4.8 / Sonnet 4.6 | Claude Sonnet + GPT-4o | GPT-4o + Codex |
| Codebase awareness | 1M context, reads all files | .cursorrules context | Repository context |
| Git integration | Native: commits, PRs, branches | Built-in source control | Copilot Chat in PRs |
| Multi-file edits | Full project scope | Multi-file via inline | Context-limit aware |
| Terminal commands | Full shell access | Limited terminal pane | No terminal |
| Pricing | Pay-per-token (API) or Pro plan | $20-40/mo subscription | Free tier + $10-20/mo |
| Custom workflows | Skills + Hooks + CLAUDE.md | .cursorrules | Copilot instructions |
| CI/CD | GitHub Actions, GitLab CI/CD | None | GitHub-native |
| Sub-agents | Yes (parallel tasks) | No | No |
| Desktop app | Yes | No | No |
Best combo: Cursor for daily IDE work, Claude Code for complex refactors, Copilot for inline completions. See the Coding Tools Comparison for benchmarks.
Pricing
| Plan | What You Get | Cost |
|---|---|---|
| Claude Pro | Claude Code access on 1 surface | $20/mo |
| Claude Max | Claude Code across all surfaces + extended usage | 200/mo (team) |
| API (Console) | Pay-per-token via Anthropic Console | Opus 25, Sonnet 15, Haiku 5 per 1M |
| Enterprise | Custom plans, volume discounts | Contact sales |