Claude Directory

Claude Code setups for learning a new codebase

Land in an unfamiliar repo and ship in a week. Agents, skills, MCP servers, and how-tos that help Claude Code map architecture, surface conventions, and build the mental model you need to be useful fast — and persist what you learn to memory for the next session.

36 items across skills, agents, plugins, MCP servers, prompts, hooks, and guides.

Example prompts

Concrete things you can type into Claude Code for codebase onboarding. Each example uses a real skill, agent, slash command, hook, or MCP server listed below.

Generate a CLAUDE.md for a repo you've never seen
```
/init
```

Map the top-level architecture

Use the code-explorer agent to produce an architecture diagram and identify the top 5 modules by churn.

Find where a feature is implemented

Use the Explore subagent: 'where does auth happen and which files reference the session cookie?'

Persist what you learn across sessions

Save memories about ownership, conventions, and gotchas as you explore — they'll load on the next session.

Skills (5)

Browse all skills

Context Engineering Kit

Advanced context engineering techniques and patterns with minimal token footprint for efficient Claude Code sessions

contextoptimizationefficiencytokens

Architecture Diagram Generator

Generate Mermaid diagrams showing system architecture, data flows, and component relationships from your codebase

architecturediagramsmermaidvisualization+1

Project Environment Setup

Detect project stack and automatically configure the local development environment with dependencies, tools, and services

setupenvironmentdevtoolsonboarding+1

Code Walkthrough

Generate an interactive walkthrough of a codebase feature, tracing execution flow from entry point to output

code-reviewdocumentationonboardingarchitecture+1

Consolidate Memory

Reflective pass over your Claude Code memory files — merge duplicates, fix stale facts, prune the MEMORY.md index

memorymaintenanceofficialanthropic+1

Agents (6)

Browse all agents

Professional TypeScript engineer specializing in scalable, type-safe applications for Node.js and browser environments

Developmenttypescriptnode.js+3

Backend Architect

Senior backend engineer specializing in scalable API design, microservices, and distributed systems

Developmentbackendapi+3

Documentation Expert

Technical writing specialist for creating clear, comprehensive documentation and CLAUDE.md files

Specializationdocumentationtechnical-writing+3

Deeply analyzes existing codebase features by tracing execution paths, mapping architecture layers, understanding patterns and abstractions, and documenting dependencies

Developmentanalysisarchitecture+3

Designs feature architectures by analyzing existing codebase patterns and conventions, then providing comprehensive implementation blueprints with specific files to create/modify

Developmentarchitecturedesign+3

Monorepo Architect

Specialist in designing and maintaining monorepo architectures with efficient build systems and dependency management

Infrastructuremonorepoturborepo+5

Plugins (13)

Browse all plugins

Feature Development

Guided feature development with codebase understanding and architecture focus. A comprehensive 7-phase workflow for building features systematically with specialized agents for exploration, architecture, and review.

featuredevelopmentarchitectureofficial+1

Systematic development framework transforming Claude Code into a proactive development partner with structured context management (deprecated, see PlanKit)

contextframeworkdevelopmentproductivity

Codebase Documenter

Analyze a service or codebase component and create comprehensive documentation in CLAUDE.md files

documentationreadmeclaude-mdautomation+1

Ralph Autonomous Mode

Autonomous overnight coding with fresh context management. Perfect for long-running tasks that need to maintain coherence across extended sessions

autonomousovernightcontextlong-running+1

Upstash Context7 MCP server for up-to-date documentation lookup. Pull version-specific documentation and code examples directly from source repositories into your LLM context. Never work with outdated docs again.

documentationcontextupstashofficial

AI code review agent for GitHub and GitLab. View and resolve Greptile's PR review comments directly from Claude Code. Natural language codebase search and intelligent code analysis.

code-reviewaigithubgitlab+2

Cursor Rules Converter

Import and convert .cursorrules files to CLAUDE.md format for seamless migration from Cursor to Claude Code

migrationcursorclaude-mdconversion+1

Persistent memory system that captures coding session context, compresses observations with AI, and injects relevant history into future sessions with SQLite storage

memorycontextpersistencesqlite+1

Portable memory in a single .mv2 file with no database dependencies. Rust-based core with sub-millisecond operations, git-committable and shareable

memoryportablerustcontext+1

Long-term learning system that observes your activity over time and automatically updates CLAUDE.md with accumulated learnings about preferences and patterns

memorylearningclaude-mdautomation+1

Architecture Patterns

Implement proven backend architecture patterns including Clean Architecture, Hexagonal Architecture, and Domain-Driven Design. Use when architecting complex backend systems or refactoring existing applications for better maintainability.

architecturegoai

Health Check Endpoints

Health check endpoints for liveness, readiness, dependency monitoring. Use for Kubernetes, load balancers, auto-scaling, or encountering probe failures, startup delays, dependency checks, timeout configuration errors.

architecturekubernetesmonitoringai

Microservices Patterns

Design microservices architectures with service boundaries, event-driven communication, and resilience patterns. Use when building distributed systems, decomposing monoliths, or implementing microservices.

MCP Servers (12)

Browse all MCP servers

Brave Search Server

Web and local search capabilities using the Brave Search API

searchwebbraveofficial

Persistent memory storage using a knowledge graph for long-term context retention

memoryknowledge-graphpersistenceofficial

Official Redis MCP server for data storage, caching, and search operations through natural language

databaserediscachesearch+1

AI-optimized search engine for agents with intelligent result extraction and summarization

searchwebairesearch+1

Neural search engine designed for AI that understands meaning and finds precisely relevant content

searchwebaisemantic+1

Elasticsearch Server

Search, index, and analyze data in Elasticsearch clusters with natural language queries

searchelasticsearchanalyticslogging+1

Context7 Server

Provide up-to-date, version-specific library documentation directly in your prompts instead of outdated training data

documentationcontextlibrariescommunity

Pinecone Server

Manage vector databases, upsert embeddings, and perform similarity searches with Pinecone

pineconevector-databaseembeddingsai+2

Connect to Weaviate vector database for semantic search, knowledge base management, and RAG workflows with vector embeddings

weaviatevector-databaseembeddingsai+2

Manage Milvus vector database collections, perform similarity searches, and handle vector embeddings for AI applications

milvusvector-databaseembeddingsai+1

Interact with ChromaDB for semantic document search, metadata filtering, and persistent vector storage in RAG pipelines

chromadbvector-databaseembeddingsai+2

Vector search engine acting as a semantic memory layer for storing and retrieving information using natural language

vector-databaseqdrantembeddingssearch+1

Other use cases

Code Review Testing Security Git Workflows Documentation Debugging Performance Deployment & CI/CD Databases API Development Refactoring Frontend Development AI & Agent Development Mobile Development Observability & Monitoring Refactoring Frontend Development