Claude Code, explained from the codebase outward
This hub translates the local Claude Code source snapshot into practical guidance for developers: how the agent loop works, how context and memory are assembled, where permissions live, how MCP fits in, and why sessions and compaction matter for real-world workflows.
Phase 1 reference guides
Readable summaries, not source dumps
Local implementation baseline
Ecosystem registry preview
Phase 3 extends the hub into reusable assets such as MCP stacks, memory templates, command packs, hook bundles, and agent specs.
Starter MCP Stack
A practical default external capability stack for everyday Claude Code work
A starter bundle that covers filesystem access, repository context, docs lookup, and persistent knowledge without assuming a huge custom setup.
Monorepo Command Pack
A command pack for onboarding, scoped changes, and targeted verification in large repos
A practical command collection for large repositories where context discipline and narrow verification matter more than raw generation speed.
Team Memory Template
A persistent memory layout for teams that want reusable project context
A file-based memory structure that separates user preferences, project context, warnings, and reference material so future conversations can recall the right things cleanly.
System map
The codebase points to a simple operating model: context and memory feed the agent loop, permissions and tools constrain execution, MCP expands the capability surface, and sessions keep long-running work durable.
Context
CLAUDE.md, git snapshot, date, instructions
Agent Loop
Query, tool calls, follow-up reasoning, compaction
Permissions
Hooks, classifiers, dialogs, mode-sensitive rules
Integrations
MCP servers, plugins, external capabilities
Durability
Transcripts, resume, memory recall, long sessions
Core deep dives
Start here for the subsystems that explain how Claude Code actually operates.
How Claude Code Works
The high-level architecture behind a Claude Code turn
A code-backed explanation of how Claude Code assembles context, runs the agent loop, invokes tools, and keeps long sessions moving.
Context and Memory
How Claude Code builds working context and persistent memory
Understand the difference between session context, CLAUDE.md instructions, MEMORY.md indexing, and relevance-based memory recall.
Permissions and Tools
How Claude Code decides what it can do and when it should ask
A practical guide to built-in tools, permission checks, and why tool availability changes across modes and agent types.
Agents and Plan Mode
How Claude Code separates planning, coordination, and delegated execution
Understand the difference between the main thread, plan mode, coordinator behavior, and constrained sub-agents.
MCP and Plugins
How Claude Code connects to external servers, tools, and resources
A map of MCP connection management, plugin-facing flows, and how external capabilities become part of the Claude Code experience.
Sessions, Resume, and Compaction
How Claude Code persists work across long conversations
A guide to transcript storage, session resume behavior, cross-project boundaries, and why long-lived conversations need compaction.
Workflow guides
These pages turn the internals into practical working patterns for debugging, review, onboarding, and long sessions.
Debugging Workflow
A practical Claude Code path for finding, reproducing, and fixing bugs
Use Claude Code effectively for debugging by combining context assembly, safe inspection tools, targeted reproduction, and focused fix loops.
Code Review Workflow
How to use Claude Code for high-signal review instead of generic commentary
A review workflow built around bug finding, risk analysis, and missing-test detection rather than vague style feedback.
Repo Onboarding Workflow
How to bring Claude Code into a new codebase without losing the plot
A practical onboarding path for large or unfamiliar repositories, built around context capture, file discovery, and a stable working plan.
Managing Long Session Context
How to avoid context drift in extended Claude Code sessions
A guide to keeping long-running work coherent by using plans, memory correctly, and prompts that resist context sprawl.
Comparisons
Use these when you need sharper boundaries between similar concepts.
Claude Code vs Cursor
A systems-level comparison of two different AI coding product shapes
Compare Claude Code and Cursor through workflow boundaries, tool models, session behavior, and integration philosophy rather than vague feature checklists.
CLAUDE.md vs MEMORY.md
Two persistence layers with different jobs
A clear comparison between repository-scoped collaboration instructions and persistent memory files intended to survive across future conversations.
Plan Mode vs Direct Execution
When Claude Code should align first and when it should just work
Use plan mode for non-trivial tasks that need decision clarity; use direct execution for bounded requests where the path is already clear.
MCP vs Built-in Tools
Choosing between native Claude Code capabilities and external capability layers
Built-in tools handle core local work; MCP expands Claude Code into connected systems, external resources, and organization-specific workflows.
Reference
Reference pages collect commands and terminology in one place.
Commands Reference
The command families that shape the Claude Code user experience
A practical index of the command groups exposed by Claude Code and what each one is for.
Glossary
A compact map of the terms that show up across the Claude Code hub
Quick definitions for the concepts that matter most across planning, sessions, memory, tools, MCP, and delegated work.
Suggested order after Phase 1
The next useful reading order is workflows first, comparisons second, then concept graph and diagrams. That matches the most natural expansion path for the current app architecture.
Workflows first
Turn the internals into practical debugging, review, onboarding, and long-context playbooks.
Comparisons next
Use sharper comparisons to reduce confusion between adjacent concepts and tools.
Then reference and maps
Once the workflows are stable, reinforce them with glossary pages and system maps.
Suggested paths
Pick the path that matches what you are trying to understand right now.
New to Claude Code
Start with the architecture overview, then read context and memory so the rest of the system makes sense.
Trying to work safely
Understand where tool permissions come from, how planning differs from execution, and how agent delegation is constrained.
Optimizing real workflows
Focus on sessions, MCP, and the command surface to understand how long-running work and integrations fit together.
Related directory resources
The hub is meant to connect concepts to real tools, docs, and integrations already listed in the directory.
Claude Code Getting Started Guide
Official getting started documentation for Claude Code
Comprehensive getting started guide for Claude Code. Covers installation, configuration, basic commands, MCP setup, and best practices for new users. The essential first read for anyone starting with Claude Code.
Claude Code Best Practices
Community-driven best practices for Claude Code
Collection of best practices learned from real-world Claude Code usage. Covers CLAUDE.md optimization, effective prompting, workflow organization, and common pitfalls to avoid. Continuously updated by the community.
MCP Protocol Documentation
Complete Model Context Protocol specification
Official documentation for the Model Context Protocol (MCP). Covers the protocol specification, server implementation, tool definitions, and integration patterns. Essential reading for MCP server developers.
Filesystem MCP Server
Secure file system access with configurable permissions and sandboxing
Official MCP server providing controlled filesystem access. Read, write, and manage files within specified directories. Features include path sandboxing, configurable permissions, and support for common file operations. Perfect for local development workflows.
Context7 MCP Server
Real-time documentation lookup for libraries and frameworks
Community MCP server that provides up-to-date documentation for popular libraries and frameworks. Instead of relying on potentially outdated training data, fetch current documentation directly. Supports npm packages, Python libraries, and more.
VS Code Claude Extension
Integrate Claude Code directly into VS Code
Official VS Code extension bringing Claude Code capabilities into your editor. Features include inline assistance, file context awareness, and seamless integration with VS Code's interface. Perfect for developers who prefer IDE workflows.