Claude Code
Agentic coding tool from Anthropic that lives in your terminal, understands your codebase, and helps you code faster through natural language commands.
At a Glance
Pricing
Engagement
Available On
Developer
Updated Mar 2026
About Claude Code
Claude Code is an agentic coding tool from Anthropic that lives in your terminal, understands your codebase, and helps you code faster through natural language commands. It can read files, execute shell commands, edit code, manage git workflows, and even spawn sub-agents for parallel tasks—all while keeping you in control.
Unlike traditional code completion tools, Claude Code operates as an autonomous agent that explores your project context, takes multi-step actions, and verifies its own work. It follows a Unix philosophy: composable, scriptable, and designed to integrate with your existing CLI tools rather than replace them.
-
Agentic Search automatically indexes and understands your entire codebase without manual context selection, enabling Claude to find relevant files, trace dependencies, and answer architectural questions.
-
Direct File Operations allow Claude to read, write, and edit files in your project, displaying proposed changes as diffs for review before applying them.
-
Shell Command Execution gives Claude access to your terminal to run build commands, tests, linters, git operations, and any CLI tools you already use.
-
Git Workflow Automation handles commits, branch management, rebases, and can create pull requests directly through the GitHub CLI integration.
-
Sub-agents and Parallel Tasks let you spawn multiple Claude instances to work on different parts of a problem simultaneously, with each sub-agent maintaining its own context.
-
Model Context Protocol (MCP) connects Claude to external tools like databases, APIs, Jira, Slack, and custom services through a standardized integration layer.
-
IDE Integration provides native extensions for VS Code (including Cursor and Windsurf) and JetBrains IDEs, displaying Claude's edits inline in your editor with visual diffs.
-
GitHub Actions Integration enables automated code reviews, issue triage, and PR creation by tagging
@claudein GitHub issues and pull requests. -
Headless/Print Mode allows scripting and CI/CD integration with JSON output, making Claude programmable for automation pipelines.
-
Custom Slash Commands let you define reusable prompts as markdown files that become available through the
/menu, sharable across your team via git. -
CLAUDE.md Context Files provide persistent project context, coding standards, and instructions that Claude reads at the start of each session.
-
Checkpointing and Rewind save conversation state so you can roll back to any previous point if Claude goes in an unexpected direction.
To get started, install Claude Code via curl -fsSL https://claude.ai/install.sh | bash (macOS/Linux) or irm https://claude.ai/install.ps1 | iex (Windows), navigate to your project directory, and run claude. Claude Code requires a Claude subscription (Pro, Max, Team, or Enterprise) or API access through the Claude Console.
Slash Commands Reference
| Command | Description |
|---|---|
/help | Show all available commands including custom project commands |
/model | Switch between Claude models (Opus, Sonnet, Haiku) |
/config | Open settings interface for model, permissions, and preferences |
/permissions | View and update tool permissions |
/allowed-tools | Manage which tools Claude can use |
/clear | Clear conversation history and start fresh |
/compact | Compress conversation to save context tokens |
/context | Monitor current token usage |
/rewind | Roll back to a previous message in the session |
/export | Export conversation to file or clipboard |
/bug | Report issues directly to Anthropic |
/ide | Connect Claude Code to your IDE |
/terminal-setup | Configure terminal for optimal Claude Code experience |
/install-github-app | Set up GitHub integration for PR reviews and @claude tagging |
/hooks | Configure pre/post tool execution hooks |
/agents | View and manage sub-agents |
/mcp | Manage MCP server connections |
/commit | Stage changes and create a commit with AI-generated message |
/review | Request a code review of current changes |
/project:* | Run custom project commands from .claude/commands/ |
CLI Flags Reference
| Flag | Description |
|---|---|
-p, --print | Non-interactive mode; run query and exit (headless) |
-c, --continue | Continue the most recent conversation |
--resume <id> | Resume a specific session by ID |
--model <model> | Specify which Claude model to use |
--output-format json | Output results as JSON with metadata |
--output-format stream-json | Stream JSON messages in real-time |
--allowedTools <list> | Restrict which tools Claude can use |
--disallowedTools <list> | Explicitly deny specific tools |
--append-system-prompt | Add instructions while keeping defaults |
--system-prompt | Fully replace the default system prompt |
--permission-mode | Set permission handling (default, acceptEdits, etc.) |
--mcp-config <file> | Load MCP server configuration |
--mcp-debug | Enable MCP debugging output |
--debug | Enable verbose debug logging |
--verbose | Show detailed execution information |
Core Tools
| Tool | Description |
|---|---|
Read | Read file contents |
Write | Create or overwrite files |
Edit | Make targeted edits to existing files |
Grep | Search file contents with patterns |
Glob | Find files matching patterns |
Bash | Execute shell commands |
mcp__* | Access MCP server tools (e.g., mcp__github__create_pull_request) |
Configuration Files
| File | Scope | Purpose |
|---|---|---|
~/.claude/settings.json | Global | User-wide settings for all projects |
.claude/settings.json | Project | Team settings, checked into git |
.claude/settings.local.json | Project | Personal settings, git-ignored |
~/.claude/CLAUDE.md | Global | Instructions for all projects |
CLAUDE.md | Project | Project-specific context and standards |
~/.claude/commands/*.md | Global | Personal custom commands |
.claude/commands/*.md | Project | Team custom commands |
.claude/agents/*.md | Project | Custom sub-agent definitions |
Platform Availability
| Platform | Method |
|---|---|
| Terminal (macOS/Linux/Windows) | Native CLI via claude command |
| VS Code | Official extension from Anthropic |
| Cursor / Windsurf | Same VS Code extension |
| JetBrains IDEs | Beta plugin (IntelliJ, PyCharm, WebStorm, etc.) |
| Neovim | Community plugin (claudecode.nvim) |
| GitHub | GitHub App + Actions for @claude mentions |
| Web | Browser interface at claude.ai/code |
Pricing & Access
Claude Code requires one of:
- Claude Pro ($17-20/mo) — Includes Claude Code access
- Claude Max ($100+/mo) — Higher limits, includes Claude Code
- Claude Team ($25-150/seat/mo) — Premium seat includes Claude Code
- Claude Enterprise (Contact Sales) — Premium seats include Claude Code
- API Access — Pay-as-you-go via Anthropic Console
Note: Claude Code usage can consume tokens quickly—average spend is ~$6/day for active users, compared to $20/month for some IDE alternatives.
Key Integrations
- GitHub — PR reviews, issue triage, @claude mentions, Actions workflows
- Git — Commits, branches, rebases, diffs, worktrees
- MCP Servers — Jira, Slack, Google Drive, Puppeteer, databases, custom tools
- CI/CD — Headless mode for pipelines, JSON output for automation
- VS Code / JetBrains — Inline diffs, diagnostics sharing, file references
Community Discussions
Inside Look at Using Claude Code Remote Control
Anthropic's Remote Control, currently in research preview, adds a third option for Claude Code users who need to step away mid-session. Your session keeps running on your machine while your phone becomes a window into it. The implementation details are worth understanding. Remote Control is a relay …
Claude double-usage promotion during non peak hours: March 13–27, 2026
Anthropic is running a two-week promotion that doubles your Claude usage limits during off-peak hours. It started March 13 and runs through March 27. The catch: “off-peak” means anything outside 5 AM to 11 AM Pacific (8 AM to 2 PM Eastern) on weekdays. So if you’re a night owl, early riser, or based…
Claude Opus 4.6 for Developers: What's Better, What's Worse, What Broke
Opus 4.6 is an agent-first release. The reasoning is deeper, the context window is bigger, and the model is meaningfully better at navigating codebases, logs, and browser tasks autonomously. Same pricing as 4.5. The catch: it's also more autopilot. Opus 4.6 will take actions without asking, fabricat…
Claude Code 2.1.27 Freezing and CPU Issues. Heres What We Know So Far
If youve updated Claude Code to version 2.1.27 and noticed things getting weird, youre not alone. Theres been a wave of reports on Reddit about this version freezing up, spiking to 100% CPU, and slowly eating through memory until everything grinds to a halt. The quick fix that's working for most peo…
Claude Code goes full on G.A. in VS Code
Claude Code is now out as a VS Code extension and its pretty wild how easy it makes things. You just click the Spark icon in your sidebar and you can start prompting Claude right there in your editor. The cool part is you can watch it work in real time. It'll analyze your code, suggest changes, and …
Pricing
Free Plan Available
Basic access to Claude
- Access to Claude
- Basic usage limits
- Chat history
Pro
Research, code, and organize
- Everything in Free
- More usage than Free
- Access more Claude models
- Unlimited Projects to organize chats
- Unlock deep Research tools
- Use extended thinking for complex work
- Connect Google Workspace: email, calendar, and docs
- Integrations with remote MCP
- Includes Claude Code
Max 5x
5x more usage than Pro with priority access and early feature access.
- Everything in Pro
- 5x more usage than Pro
- Higher output limits for all tasks
- Early access to advanced Claude features
- Priority access at high traffic times
- Includes Claude Code
Max 20x
20x more usage than Pro — 50% cheaper per unit of usage than Max 5x.
- Everything in Pro
- 20x more usage than Pro
- Higher output limits for all tasks
- Early access to advanced Claude features
- Priority access at high traffic times
- Includes Claude Code
Team - Standard Seat
For collaboration across organizations
- Everything in Pro
- More usage
- Central billing and administration
- Single sign-on (SSO) and domain capture
- Early access to collaboration features
- Chat, projects, and more
- Minimum 5 members
Team - Premium Seat
Team seat with Claude Code included
- Everything in Team Standard
- Includes Claude Code
- Minimum 5 members
Enterprise
For businesses operating at scale
- Everything in Team
- More usage
- Enhanced context window
- Role-based access with fine-grained permissioning
- System for Cross-domain Identity Management (SCIM)
- Audit logs
- Google Docs cataloging
- Claude Code available with premium seat
- Compliance API for observability and monitoring
Capabilities
Key Features
- Natural language code editing and generation
- Multi-file codebase understanding
- Agentic search and indexing
- Git workflow automation
- Test execution and debugging
- Architecture and code explanation
- Dependency analysis and management
- Shell command execution
- Extended thinking for complex reasoning
- Sub-agents for parallel task execution
- Checkpointing and rewind
- Custom slash commands
- CLAUDE.md context files
- Pre and post tool hooks
- Headless mode for scripting and CI/CD
- Background task execution
- GitHub PR and issue automation
- IDE inline diff viewing
- Context management and compaction
- File search and pattern matching
- Memory and session persistence
- Plugin and marketplace support
- Custom agent creation
- Web browser automation
- Image and screenshot analysis