Claude Code Complete Guide
From setup and commands to advanced workflows, agents, skills, and pro tips — the all-in-one reference for Claude Code.
- Source repo
- zebbern/claude-code-guide
- Stars
- ★ 4.7k
- Last updated
- today
- License
- MIT
- Primary language
- Python
- FA score
- 29/100 · Major gaps
At a glance
- How it runs
- Works with
- Platform-specificClaude CodeClaude.ai (Partial support)
- Cost
- Free software; you pay for model usage
- Setup effort
- Low · running in minutes
- You'll need
- Typical use
- A beginner setting up Claude Code for the first time can follow the Quick Start section to install and configure it on Windows, macOS, or Linux.
- Not a fit if
- Teams standardized on other AI coding agents, not Claude Code
- Anyone without a paid Claude Pro/Max/Team/Enterprise or Console account
- Source review
- 29/100 · Major gaps 8 safety controls not found
What does this agent do, and when should you use it?
This community-maintained guide covers the full Claude Code toolchain, from installation to advanced usage. It includes native and npm install methods, environment variables and configuration files (CLAUDE.md, .claude/rules), a slash-command and CLI flag cheat sheet, keyboard shortcuts and Vim mode, MCP and Hooks integration, security and permissions settings, plus automation (PR review, issue triage) and troubleshooting. The manual uses tables and copy-paste command examples for fast lookup. It targets users of Anthropic's Claude Code CLI, helping them code more effectively in the terminal. The repository also references official docs and other ecosystem resources (e.g., Discord integration, skills collections), but contains no executable agent code itself; it relies entirely on the Anthropic Claude Code environment.
This repository is a documentation project, not runnable code, but it explains Claude Code's full workflow. It teaches how to install Claude Code via the native installer (curl scripts) or npm, how to set ANTHROPIC_API_KEY or authenticate with claude auth login, and lists every slash command (e.g., /init, /plan, /mcp, /agents, /compact) and CLI flag (e.g., -p, --output-format json, --allowedTools). It explains the four memory layers (enterprise CLAUDE.md, project CLAUDE.md, user ~/.claude/CLAUDE.md, and local CLAUDE.local.md) and their loading order, and covers MCP server configuration (claude mcp add/list) and the Hooks system. It also covers automation scenarios like using claude -p --json-schema for structured output, running background sessions with --bg, and using --auto-pr-review for PR review.
- A beginner setting up Claude Code for the first time can follow the Quick Start section to install and configure it on Windows, macOS, or Linux.
- A developer wants to enforce project conventions by defining architecture and coding standards in CLAUDE.md and generating it with /init.
- A security-conscious user needs to restrict Claude's file access using --allowedTools, --permission-mode plan, or a security manager (e.g., security audit policy).
- A team wants to integrate external services like Discord or GitHub via MCP servers, following the MCP section with
claude mcp add. - A developer reviews code in bulk using /code-review --comment for PR comments or /ultrareview for comprehensive cloud review.
- An organization standardizes development rules by modularizing them in .claude/rules/ and sharing via source control.
How do you install or deploy this agent?
This repository is documentation; no installation is needed. To use Claude Code itself, follow its installation methods:
- macOS/Linux: curl -fsSL https://claude.ai/install.sh | bash
- Windows (PowerShell): irm https://claude.ai/install.ps1 | iex- Or via npm: npm install -g @anthropic-ai/claude-code (requires Node.js 18+).
After installing, verify with claude --version.
How do you use this agent?
- Run
claudein your project directory to start an interactive session. - On first use, run
claude auth loginto authenticate your Anthropic account, or set the ANTHROPIC_API_KEY environment variable. - Configure environment variables as needed: e.g.,
export ANTHROPIC_MODEL="sonnet"orexport CLAUDE_CODE_USE_BEDROCK=1. - Create a CLAUDE.md file with project instructions, or use /init to generate it automatically.
- Use /plan, /compact, /mcp, etc. to manage sessions.
- Consult the Cheat Sheet and Slash Command tables in this guide for reference.
What are this agent's strengths and limitations?
- Covers the full spectrum of Claude Code features, from installation to advanced capabilities like subagents, skills, and background tasks.
- Provides countless copy-paste command examples, reducing trial-and-error and speeding up learning.
- Explains environment variables and the memory hierarchy of configuration files for consistent cross-project behavior.
- Includes MCP integration and Hooks extension mechanisms for expanding Claude Code's capabilities.
- Exclusively targets Claude Code, not other AI coding agents, creating strong vendor lock-in.
- Static documentation depends on Anthropic's updates and may lag behind the latest CLI changes.
- No actual runnable agent code; it can't be directly integrated with other systems without the official CLI.
- Some advanced features (like Agent Teams) require extra configuration or experimental toggles and may need manual testing.
How does this agent compare with similar options?
Key facts side by side with the most closely related agents.
| Agent | Source review | Form / cost | Stars | Updated | Language | Full support on |
|---|---|---|---|---|---|---|
| Claude Code Complete Guide This agent | 29 · Major gaps | CLIFree + model costs | ★ 4.7k | today | Python | Claude Code |
| ZCF - Zero-Config Code Flow | 51 · Major gaps | CLIFree + model costs | ★ 6.1k | 1mo ago | TypeScript | Codex · Claude Code |
| Endoplexity Browser Agent | 67 · Some gaps | Browser extensionPaid | ★ 180 | 2mo ago | TypeScript | Claude.ai |
| Tokless | 57 · Major gaps | CLIFree | ★ 293 | today | Go | Codex · Claude Code |
How does FollowAgents rate this agent?
Why each dimension lost points
Evidence shows this is a documentation repository with no executable code or permission configurations. Therefore, all trust-related criteria (least privilege, user confirmation, data flow transparency, sensitive data handling, dependency security, external effects, rollback, source attribution) score 0. There is no evidence of malicious behavior, but also no security mechanisms.
Self-consistency: README is well-structured, but some content (e.g., environment variable list) may be incomplete, and no verification method is provided, score 1. Dependency availability: Depends on external services (Anthropic API, GitHub), but no availability guarantees are provided, score 1. Failure messages: No detailed error handling or troubleshooting instructions, score 0.
Audience and scenarios: Clearly targets Claude Code users from beginners to power users, covering multiple scenarios, score 2. Capability boundaries: Does not clearly state the scope or limitations of the documentation, score 1. Trigger precision: Not applicable, score 0. Environment fit: Provides installation and configuration instructions for multiple platforms (Windows, macOS, Linux), score 2.
Information architecture: Clear table of contents and well-organized content, score 2. Install notes: Provides multiple installation methods, score 2. Naming stability: No mention of naming conventions, score 1. Examples and FAQ: Provides numerous command examples but lacks FAQ, score 2. Known limitations: Not mentioned, score 0. License: MIT license, score 2. Versioning and changelog: Has CHANGELOG.md but no versioning policy, score 1. Maintenance responsibility: Has GitHub Actions for auto-sync but no explicit maintainer, score 1.
Output usability: Provides many directly usable commands and configurations, score 2. Marginal value: As a community guide, provides practical information beyond official docs, score 2. Cost-benefit: Free and open source, but requires user evaluation, score 1.
Claim traceability: Some content references official docs but lacks specific links, score 1. Cross-source corroboration: No verification from multiple sources, score 1. Fact-inference separation: Does not clearly distinguish facts from inferences, score 1.
- Not found in source: least-privilege scopingGrant only what the task needs: a dedicated account or read-only token, scoped to specific directories and repos.
- Not found in source: confirmation before actingTurn on (or add) a confirmation step before it acts, and try it in a sandbox or test environment before real data.
- Not found in source: data-flow disclosureWatch which external services it contacts (proxy or firewall logs) and keep sensitive data out until you know where it goes.
- Not found in source: sensitive-data handlingUse dedicated, low-privilege, revocable API keys — never production credentials — and keep secrets out of logs.
- Not found in source: dependency securityPin versions and run a dependency audit (npm audit, pip-audit) before installing; prefer running it in a container.
- Not found in source: disclosed external effectsEstablish which external systems it writes to, sends to or changes, and verify with test accounts or repos before production.
- Not found in source: rollback or recovery pathBack up first, or work on a git branch or snapshot, so its changes can be undone.
- Not found in source: verifiable attributionInstall from the official repo or registry and check the publisher and URL to avoid look-alike packages.
- This repository is documentation-only, with no executable code, so runtime security and reliability cannot be assessed.
- Environment variables and configuration examples may be outdated or inaccurate; users should verify against official documentation.
- The auto-sync workflow pulls content from external sources, posing supply chain risks, but no malicious behavior was found.
FAQ
Is this repository a copy of the official documentation?
Do I need a paid subscription to use Claude Code?
How can I ensure my code's security?
What if I encounter MCP server connection issues?
claude mcp list for diagnosis.