Skip to content

basic-usage.md

Latest commit

 

History

History
152 lines (113 loc) · 2.41 KB

basic-usage.md

File metadata and controls

152 lines (113 loc) · 2.41 KB

Basic Usage Examples

Common ClaudeForge usage scenarios.

Example 1: New TypeScript React Project

Scenario: Starting fresh project, need CLAUDE.md

Setup:

mkdir my-react-app
cd my-react-app
npm init -y
npm install react react-dom typescript

Run ClaudeForge:

/enhance-claude-md

Claude's Response:

Discovered:
- Project Type: Web App
- Tech Stack: TypeScript, React, Node.js
- Team Size: Solo
- Phase: Prototype

Creating CLAUDE.md (85 lines) with:
- Project structure diagram
- Setup instructions (npm install, npm start)
- Component guidelines
- TypeScript best practices

Output: Single CLAUDE.md file, ~85 lines

Example 2: Python FastAPI Project

Scenario: API service, team of 6

Setup:

mkdir api-service
cd api-service
echo "fastapi[all]" > requirements.txt
echo "pytest" >> requirements.txt
mkdir app tests

Run ClaudeForge:

/enhance-claude-md

"This is a Python FastAPI API service. Team of 6 developers, MVP phase."

Output: CLAUDE.md with:

  • FastAPI patterns
  • Async/await guidelines
  • Testing with pytest
  • API documentation standards

Example 3: Full-Stack Application

Scenario: Large project with backend + frontend

Setup:

mkdir fullstack-app
cd fullstack-app
mkdir backend frontend
cd backend && echo "express" > package.json
cd ../frontend && echo "react" > package.json

Run ClaudeForge:

/enhance-claude-md

Output: Modular structure

  • CLAUDE.md (root navigation)
  • backend/CLAUDE.md (API guidelines)
  • frontend/CLAUDE.md (React patterns)

Example 4: Enhance Existing Basic File

Before:

# CLAUDE.md

## Tech Stack
- TypeScript
- React

Run:

/enhance-claude-md

Claude Analyzes:

Quality Score: 25/100

Missing:
- Project Structure
- Setup & Installation
- Core Principles
- Common Commands

After: Complete 120-line file with all sections

Example 5: Quality Check Before Commit

Scenario: Made edits, verify quality

Run:

/enhance-claude-md

"Just validate, don't make changes"

Output:

Quality Score: 88/100

✅ Length: 245 lines (good)
✅ Structure: All required sections
✅ Formatting: Valid markdown
⚠️  Consider adding: Performance Guidelines

See also: