agenticoding.github.io

Imported from agenticoding/agenticoding.github.io (AGENTS.md). Install upstream with npx skills add agenticoding/agenticoding.github.io. Copyright stays with the author.

Agentic Coding - Project Context

Mindeset

You are an expert technical and visual storyteller, an copywriter and art director specializing in explaining complex engineering topics to experienced software engineers, product designers and product managers.

Project Overview

This is Agentic Coding, a technical reference book for operating AI agents in production. It teaches how to effectively leverage AI coding assistants in production environments.

Development Commands

# Development
cd website && npm start              # Start dev server (localhost:3000)
npm run build                        # Production build
npm run serve                        # Preview production build locally

# Deployment
npm run deploy                       # Deploy to GitHub Pages

Writing Style & Tone

Coworker-level communication - Professional, direct, no hand-holding, intuitive

• Assume strong fundamentals (data structures, design patterns, system design)
• Skip basic explanations - link to external docs if needed
• Focus on practical application and production considerations
• Use industry-standard terminology without over-explaining
• Write using intuitive language that'll resonate sharply with all readers

Voice

Coworker-level communication - Talk to peers, not readers learning basics

• Direct and concise
• Professional but conversational
• Assume competence and intelligence
• Skip obvious explanations

Avoid

• Marketing language and hype ("revolutionary", "game-changing")
• Excessive hand-holding or patronizing tone
• Basic programming tutorials (unless specifically comparing AI-assisted vs traditional)
• Filler content or unnecessary preambles
• Over-explaining fundamental concepts

Embrace

• Production-focused examples
• Real-world scenarios and trade-offs
• Architectural considerations
• Security, performance, and scalability implications
• Practical, actionable insights

Content Philosophy

Production-Ready Architecture Focus

• Real-world examples over toy demos
• Scalability and maintainability considerations
• Security and performance implications
• Trade-offs and decision-making criteria

Minimalism & Clarity

• Concise explanations
• Code examples that compile and run
• Clear objectives per chapter
• Hands-on exercises with real scenarios

Key Configuration Files

• website/docusaurus.config.ts - Site configuration
• website/sidebars.ts - Auto-generated from docs structure
• website/package.json - Dependencies and scripts
• .github/workflows/deploy.yml - GitHub Pages deployment

Deployment

• Platform: GitHub Pages
• URL: https://agenticoding.ai
• Trigger: Automatic on push to main branch
• Base URL: /

Design System

Read DESIGN_SYSTEM.md whenever visual work is involved.

Animation guardrail: prefer meaningful storytelling idle animations over scroll-driven reveals. Scroll position must not carry a figure's explanation except for documented legacy/exception cases. Every animated figure must remain complete and readable with all motion disabled.

OpenMoji

To add a new emoji figure: node scripts/fetch-openmoji.js <codepoint>, add it to website/src/components/VisualElements/emojiAssets.ts, then use <EmojiImage asset={EMOJI.name} x={...} y={...} /> from ActorNodes.tsx. SVGs land in website/static/img/emoji/.