Claude Code Ultimate Guide

6 months of daily practice distilled into a guide that teaches you the WHY, not just the what. From core concepts to production security, you learn to design your own agentic workflows instead of copy-pasting configs.

If this guide helps you, give it a star ⭐ — it helps others discover it too.

StarMapper

Choose Your Path

🎯 What You'll Learn

This guide teaches you to think differently about AI-assisted development:

• ✅ Understand trade-offs — When to use agents vs skills vs commands (not just how to configure them)
• ✅ Build mental models — How Claude Code works internally (architecture, context flow, tool orchestration)
• ✅ Visualize concepts — 41 Mermaid diagrams covering model selection, master loop, memory hierarchy, multi-agent patterns, security threats, AI fluency paths
• ✅ Master methodologies — TDD, SDD, BDD with AI collaboration (not just templates)
• ✅ Security mindset — Threat modeling for AI systems (only guide with 24 CVEs + 655 malicious skills database)
• ✅ Test your knowledge — 271-question quiz to validate understanding (no other resource offers this)

Outcome: Go from copy-pasting configs to designing your own agentic workflows with confidence.

📊 When to Use This Guide vs Everything-CC

Both guides serve different needs. Choose based on your priority.

Ecosystem Positioning

                    EDUCATIONAL DEPTH
                           ▲
                           │
                           │  ★ This Guide
                           │  Security + Methodologies + 24K+ lines
                           │
                           │  [Everything-You-Need-to-Know]
                           │  SDLC/BMAD beginner
  ─────────────────────────┼─────────────────────────► READY-TO-USE
  [awesome-claude-code]    │            [everything-claude-code]
  (discovery, curation)    │            (plugin, 1-cmd install)
                           │
                           │  [claude-code-studio]
                           │  Context management
                           │
                      SPECIALIZED

4 unique gaps no competitor covers:

1. Security-First — 24 CVEs + 655 malicious skills tracked (no competitor has this depth)
2. Methodology Workflows — TDD/SDD/BDD comparison + step-by-step guides
3. Comprehensive Reference — 24K+ lines across 16 specialized guides (24× more reference material than everything-cc)
4. Educational Progression — 271-question quiz, beginner → expert path

Recommended workflow:

1. Learn concepts here (mental models, trade-offs, security)
2. Use battle-tested configs there (quick project setup)
3. Return here for deep dives (when something doesn't work or to design custom workflows)

Both resources are complementary, not competitive. Use what fits your current need.

⚡ Quick Start

Quickest path: Cheat Sheet — 1 printable page with daily essentials

Interactive onboarding (no setup needed):

claude "Fetch and follow the onboarding instructions from: https://raw.githubusercontent.com/FlorianBruniaux/claude-code-ultimate-guide/main/tools/onboarding-prompt.md"

Browse directly: Full Guide | Visual Diagrams | Examples | Quiz

🔌 MCP Server — Use the guide from any Claude Code session

No cloning needed. Add to ~/.claude.json and ask questions directly from any session:

{
  "mcpServers": {
    "claude-code-guide": {
      "type": "stdio",
      "command": "npx",
      "args": ["-y", "claude-code-ultimate-guide-mcp"]
    }
  }
}

17 tools: search_guide, read_section, get_cheatsheet, get_digest, get_example, list_examples, search_examples, get_release, get_changelog, compare_versions, list_topics, get_threat, list_threats, plus init_official_docs, refresh_official_docs, diff_official_docs, search_official_docs (v1.1.0 — official Anthropic docs tracker) — plus 13 slash commands /ccguide:* and a Haiku agent.

Onboarding one-liner (once MCP is configured):

claude "Use the claude-code-guide MCP server. Activate the claude-code-expert prompt, then run a personalized onboarding: ask me 3 questions about my goal, experience level, and preferred tone — then build a custom learning path using search_guide and read_section to navigate the guide with live source links."

→ MCP Server README

📁 Repository Structure

graph LR
    root[📦 Repository<br/>Root]

    root --> guide[📖 guide/<br/>24K+ lines]
    root --> examples[📋 examples/<br/>228 templates]
    root --> quiz[🧠 quiz/<br/>271 questions]
    root --> tools[🔧 tools/<br/>utils]
    root --> machine[🤖 machine-readable/<br/>AI index]
    root --> docs[📚 docs/<br/>115 evaluations]

    style root fill:#d35400,stroke:#e67e22,stroke-width:3px,color:#fff
    style guide fill:#2980b9,stroke:#3498db,stroke-width:2px,color:#fff
    style examples fill:#8e44ad,stroke:#9b59b6,stroke-width:2px,color:#fff
    style quiz fill:#d68910,stroke:#f39c12,stroke-width:2px,color:#fff
    style tools fill:#5d6d7e,stroke:#7f8c8d,stroke-width:2px,color:#fff
    style machine fill:#138d75,stroke:#16a085,stroke-width:2px,color:#fff
    style docs fill:#c0392b,stroke:#e74c3c,stroke-width:2px,color:#fff

📦 claude-code-ultimate-guide/
│
├─ 📖 guide/              Core Documentation (24K+ lines)
│  ├─ ultimate-guide.md   Complete reference, 10 sections
│  ├─ cheatsheet.md       1-page printable
│  ├─ architecture.md     How Claude Code works internally
│  ├─ methodologies.md    TDD, SDD, BDD workflows
│  ├─ diagrams/           41 Mermaid diagrams (10 thematic files)
│  ├─ third-party-tools.md  Community tools (RTK, ccusage, Entire CLI)
│  ├─ mcp-servers-ecosystem.md  Official & community MCP servers
│  └─ workflows/          Step-by-step guides
│
├─ 📋 examples/           228 Production Templates
│  ├─ agents/             9 custom AI personas
│  ├─ commands/           26 slash commands
│  ├─ hooks/              31 hooks (bash + PowerShell)
│  ├─ skills/             14 skills (9 on SkillHub)
│  └─ scripts/            Utility scripts (audit, search)
│
├─ 🧠 quiz/               271 Questions
│  ├─ 9 categories        Setup, Agents, MCP, Trust, Advanced...
│  ├─ 4 profiles          Junior, Senior, Power User, PM
│  └─ Instant feedback    Doc links + score tracking
│
├─ 🔧 tools/              Interactive Utilities
│  ├─ onboarding-prompt   Personalized guided tour
│  └─ audit-prompt        Setup audit & recommendations
│
├─ 🤖 machine-readable/   AI-Optimized Index
│  ├─ reference.yaml      Structured index (~2K tokens) — powers landing site CMD+K search
│  ├─ claude-code-releases.yaml  Structured releases changelog
│  └─ llms.txt            Standard LLM context file
│
└─ 📚 docs/               115 Resource Evaluations
   └─ resource-evaluations/  5-point scoring, source attribution

🎯 What Makes This Guide Unique

🎓 Deep Understanding Over Configuration

Outcome: Design your own workflows instead of copy-pasting blindly.

We teach how Claude Code works and why patterns matter:

• Architecture — Internal mechanics (context flow, tool orchestration, memory management)
• Trade-offs — Decision frameworks for agents vs skills vs commands
• Configuration Decision Guide — Unified "which mechanism for what?" map across all 7 config layers
• Pitfalls — Common failure modes + prevention strategies

What this means for you: Troubleshoot issues independently, optimize for your specific use case, know when to deviate from patterns.

🖼️ Visual Diagrams Series (41 Mermaid Diagrams)

Outcome: Grasp complex concepts instantly through visual mental models.

41 interactive diagrams across 10 thematic files — GitHub-native Mermaid rendering + ASCII fallback for every diagram:

• Foundations — 4-layer context model, 9-step pipeline, permission modes
• Architecture — Master loop, tool categories, system prompt assembly
• Multi-Agent — 3 topologies, worktrees, dual-instance, horizontal scaling
• Security — 3-layer defense, MCP rug pull attack chain, verification paradox
• Cost & Models — Model selection tree, token reduction pipeline

Browse all 41 diagrams →

What this means for you: Understand the master loop before reading 24K+ lines, see multi-agent topologies at a glance, share visual security threat models with your team.

🛡️ Security Threat Intelligence (Only Comprehensive Database)

Outcome: Protect production systems from AI-specific attacks.

Only guide with systematic threat tracking:

• 24 CVE-mapped vulnerabilities — Prompt injection, data exfiltration, code injection
• 655 malicious skills catalogued — Unicode injection, hidden instructions, auto-execute patterns
• Production hardening workflows — MCP vetting, injection defense, audit automation

Threat Database → | Security Guide →

What this means for you: Vet MCP servers before trusting them, detect attack patterns in configs, comply with security audits.

📝 271-Question Knowledge Validation (Unique in Ecosystem)

Outcome: Verify your understanding + identify knowledge gaps.

Only comprehensive assessment available — test across 9 categories:

• Setup & Configuration, Agents & Sub-Agents, MCP Servers, Trust & Verification, Advanced Patterns

Features: 4 skill profiles (Junior/Senior/Power User/PM), instant feedback with doc links, weak area identification

Try Quiz Online → | Run Locally

What this means for you: Know what you don't know, track learning progress, prepare for team adoption discussions.

🤖 Agent Teams Coverage (v2.1.32+ Experimental)

Outcome: Parallelize work on large codebases (Fountain: 50% faster, CRED: 2x speed).

Only comprehensive guide to Anthropic's multi-agent coordination:

• Production metrics from real companies (autonomous C compiler, 500K hours saved)
• 5 validated workflows (multi-layer review, parallel debugging, large-scale refactoring)
• Decision framework: Teams vs Multi-Instance vs Dual-Instance vs Beads

Agent Teams Workflow → | Section 9.20 →

What this means for you: Break monolithic tasks into parallelizable work, coordinate multi-file refactors, review your own AI-generated code.

🔬 Methodologies (Structured Development Workflows)

Outcome: Maintain code quality while working with AI.

Complete guides with rationale and examples:

• TDD — Test-Driven Development (Red-Green-Refactor with AI)
• SDD — Specification-Driven Development (Design before code)
• BDD — Behavior-Driven Development (User stories → tests)
• GSD — Get Shit Done (Pragmatic delivery)

What this means for you: Choose the right workflow for your team culture, integrate AI into existing processes, avoid technical debt from AI over-reliance.

📚 204 Annotated Templates

Outcome: Learn patterns, not just configs.

Educational templates with explanations:

• Agents (6), Commands (26), Hooks (31), Skills
• Comments explaining why each pattern works (not just what it does)
• Gradual complexity progression (simple → advanced)

Browse Catalog →

What this means for you: Understand the reasoning behind patterns, adapt templates to your context, create your own custom patterns.

🔍 115 Resource Evaluations

Outcome: Trust our recommendations are evidence-based.

Systematic assessment of external resources (5-point scoring):

• Articles, videos, tools, frameworks
• Honest assessments with source attribution (no marketing fluff)
• Integration recommendations with trade-offs

See Evaluations →

What this means for you: Save time vetting resources, understand limitations before adopting tools, make informed decisions.

🎯 Learning Paths

Progressive Journey

• Week 1: Foundations (install, CLAUDE.md, first agent)
• Week 2: Core Features (skills, hooks, trust calibration)
• Week 3: Advanced (MCP servers, methodologies)
• Month 2+: Production mastery (CI/CD, observability)

🔧 Rate Limits & Cost Savings

cc-copilot-bridge routes Claude Code through GitHub Copilot Pro+ for flat-rate access ($10/month instead of per-token billing).

# Install
git clone https://github.com/FlorianBruniaux/cc-copilot-bridge.git && cd cc-copilot-bridge && ./install.sh

# Use
ccc   # Copilot mode (flat $10/month)
ccd   # Direct Anthropic mode (per-token)
cco   # Offline mode (Ollama, 100% local)

Benefits: Multi-provider switching, rate limit bypass, 99%+ cost savings on heavy usage.

→ cc-copilot-bridge

🔑 Golden Rules

1. Verify Trust Before Use

Claude Code can generate 1.75x more logic errors than human-written code (ACM 2025). Every output must be verified. Use /insights commands and verify patterns through tests.

Strategy: Solo dev (verify logic + edge cases). Team (systematic peer review). Production (mandatory gating tests).

2. Never Approve MCPs from Unknown Sources

24 CVEs identified in Claude Code ecosystem. 655 malicious skills in supply chain. MCP servers can read/write your codebase.

Strategy: Systematic audit (5-min checklist). Community-vetted MCP Safe List. Vetting workflow documented in guide.

3. Context Pressure Changes Behavior

At 70% context, Claude starts losing precision. At 85%, hallucinations increase. At 90%+, responses become erratic.

Strategy: 0-50% (work freely). 50-70% (attention). 70-90% (/compact). 90%+ (/clear mandatory).

4. Start Simple, Scale Smart

Start with basic CLAUDE.md + a few commands. Test in production for 2 weeks. Add agents/skills only if need is proven.

Strategy: Phase 1 (basic). Phase 2 (commands + hooks if needed). Phase 3 (agents if multi-context). Phase 4 (MCP servers if truly required).

5. Methodologies Matter More with AI

TDD/SDD/BDD are not optional with Claude Code. AI accelerates bad code as much as good code.

Strategy: TDD (critical logic). SDD (architecture upfront). BDD (PM/dev collaboration). GSD (throwaway prototypes).

Quick Reference

Context management is critical. See the Cheat Sheet for thresholds and actions.

🤖 For AI Assistants

Quick load: curl -sL https://raw.githubusercontent.com/FlorianBruniaux/claude-code-ultimate-guide/main/machine-readable/reference.yaml

reference.yaml — Structure & Landing Site Search

reference.yaml is organized into several top-level sections:

The deep_dive section powers the landing site CMD+K search. The build script (scripts/build-guide-index.mjs) parses it to generate 160 search entries.

How the search index works

The CMD+K search on the landing site is an explicit index — not a full-text search. Only entries listed in deep_dive are indexed. Keywords are derived mechanically from the key name and file path, not from the file content.

Consequence: adding a new guide section requires explicitly adding an entry to deep_dive, then running pnpm build:search in the landing repo.

Maintaining reference.yaml

Adding a new entry to deep_dive:

deep_dive:
  # existing entries...
  my_new_section: "guide/my-new-file.md"          # local guide file
  my_hook_example: "examples/hooks/bash/foo.sh"   # example file
  my_section_ref: "guide/ultimate-guide.md:1234"  # with line number anchor

Critical: avoid duplicate keys. If a key appears twice in deep_dive, the YAML parser fails and the landing site search index becomes empty (0 entries). The build exits with a warning but no hard error:

[build-guide-index] ERROR: Failed to parse YAML: duplicated mapping key
[build-guide-index] Generating empty guide-search-entries.ts

Use distinct names — e.g. if you need both a line-number reference and a file path for the same concept, suffix the line-number key with _line:

security_gate_hook_line: 6907                              # line number ref
security_gate_hook: "examples/hooks/bash/security-gate.sh" # file path ref

📄 Whitepapers (FR + EN)

11 focused whitepapers covering Claude Code in depth — PDF + EPUB, available in French and English. 472 pages total.

Coming soon — currently in private access. Public release planned.

🗂️ Recap Cards (FR, EN coming)

57 single-page A4 reference cards — printable, one concept per card. Available in French; English version in progress.

Browse online: cc.bruniaux.com/cheatsheets/

• Technique (22 cards) — Commands, permissions, configuration, MCP, models, context window
• Méthodologie (22 cards) — Daily workflow, agents, hooks, CI/CD, multi-agent, debug
• Conception (13 cards) — Mental models, prompting, security by design, cost patterns

🌍 Ecosystem

Claude Cowork (Non-Developers)

Claude Cowork is the companion guide for non-technical users (knowledge workers, assistants, managers).

Same agentic capabilities as Claude Code, but through a visual interface with no coding required.

→ **[Claude Cowork Guide](https://github.com/FlorianBruniaux/cl