Claude Code Intro Lesson 3 of 3

Context Design

What this lesson teaches

The quality of Claude's output depends heavily on the context you provide. This lesson covers what to include, what to skip, and how to use CLAUDE.md effectively.

What Is Context?

Context is everything Claude knows during a conversation:

Files it has read
Your previous messages
Project configuration (CLAUDE.md)
Tool outputs and command results

More context isn't always better. Too much can confuse Claude and waste tokens.

CLAUDE.md: Your Project Config

Create a CLAUDE.md file in your project root. Claude reads this automatically at the start of every session.

What to include:

Project structure overview
Tech stack and conventions
Common commands (build, test, deploy)
Important patterns or rules

Example:

# My Project

## Stack
- Next.js 14 with App Router
- TypeScript strict mode
- Prisma + PostgreSQL

## Commands
- `npm run dev` - Start development
- `npm test` - Run tests
- `npm run lint` - Check code style

## Conventions
- Use server components by default
- API routes in /app/api/
- Keep components under 100 lines

Keep it short: CLAUDE.md is loaded every session. Long files waste tokens on every request.

Effective Context Strategies

Be specific about what you need

Instead of "help with auth", say "add JWT validation to the /api/users endpoint".

Point to relevant files

Instead of letting Claude search, tell it: "The user model is in src/models/user.ts".

State constraints upfront

"Use only existing dependencies" or "Keep the current API structure".

Clear conversation when switching tasks

Old context from a different task can mislead Claude. Use /clear when starting fresh.

Common mistake: Dumping entire files when only a function matters. More context = more noise = worse results.

Key Takeaways

Context quality matters more than quantity
CLAUDE.md gives Claude project knowledge automatically
Be specific about files, constraints, and goals
Clear old context when switching tasks