Coding agents
Build a practical understanding of coding agents from the underlying principles.
AUTOMATE · 35 ARTICLES
AI Coding & Agents
Move beyond chat and delegate real development work with sound command, permission, tool, context, and safety practices.
Topic map
What this topic covers
Agents and MCP
Build a practical understanding of agents and mcp from the underlying principles.
Context engineering
Build a practical understanding of context engineering from the underlying principles.
Vibe coding
Build a practical understanding of vibe coding from the underlying principles.
Start here
New here? Start with these
A guided path from fundamentals to the next step.
[Claude Code #1] Getting started with /init and automatic CLAUDE.md generation
The Claude Code /init command analyzes your codebase and automatically generates CLAUDE.md. This covers its behavior when a file already exists, migrating Cursor and Copilot rules, interactive mode, and how to refine the generated file.
Read article
[MCP·Skill #1] What Is an MCP Server? From Connection Methods to Security Considerations
Using ChatGPT or Claude can leave you wanting more. The models are smart, but they cannot see your database or read your internal wiki.
Read article
[Agent Design #1] What Is Harness Engineering? What Comes After Prompting
A harness is the execution loop wrapped around a model. We cover its components—tools, permissions, context management, and feedback—the reason agent performance is determined by the product of model intelligence and harness quality, and what makes a good harness.
Read articleAll articles
All articles
[Claude Code #3] Permissions in Review: /permissions and Mode Cycling
Claude Code permissions use allow, ask, and deny rules, evaluated with deny first. This article covers /permissions, Bash wildcard syntax, Shift+Tab mode cycling, and settings-file precedence.
Read article[Claude Code #2] /plan mode: agree on the design before making changes
Claude Code plan mode requires approval of a plan before changing code. This covers enabling it with Shift+Tab or /plan, the three approval options, editing plans with Ctrl+G, and setting defaults.
Read articleHow to prevent Vibe Coding mishaps from the start with rule files
Vibe Coding mishaps are decided when you write the first rule file, not when you read the code. Here are five rules for CLAUDE.md and .cursorrules covering security, duplication, architecture, completion criteria, and scope, with example wording.
Read article[Agent Design #3] Designing an AI Content Pipeline, from Generation to Publication
A design that splits AI content generation into independent stages—generation, review, asset creation, and publication—instead of relying on a single prompt. It covers fixed output schemas, separate review, and preserving the original even when a stage fails.
Read articleWhat Is Spec Driven Development? A Guide to spec-kit and Kiro
SDD (Spec Driven Development) is a methodology that finalizes specifications in documents before asking for code and treats them as the single source of truth. This guide covers the four-step spec·plan·tasks flow, two tools, and how it differs from vibe coding.
Read article[Agent Design #2] Context engineering: treating the context window as a budget
A larger context window is not inherently better; higher relevance is. This article summarizes four strategies—selective loading, summarization, isolation, and externalization—for using a finite window like a budget, along with design criteria for always-loaded files such as CLAUDE.md.
Read article[Vibe Coder #8] 6 security checks before accepting users
Six security checks to complete before publishing your app link. Assuming attackers cast a bot-driven net without choosing targets, this covers the doors commonly breached in vibe-coding apps: RLS, authorization checks, input validation, and more.
Read article[Open Source Finds #2] oh-my-design: Giving coding agents a brand
oh-my-design is an open-source CLI that installs a DESIGN.md design workflow in Claude Code, Codex, and Cursor. We examine 440 enterprise references, evidence practices, skill structure, and pre-writing cautions.
Read articleWhat is DESIGN.md? The file that teaches AI agents your design system
DESIGN.md is a design-system file convention for AI coding agents, open-sourced by Google Stitch. This guide covers YAML tokens, Markdown structure, the lint·export CLI, connecting it to Claude Code·Cursor, and its limitations.
Read article[Vibe Coder #7] Avoiding AI, DB, and hosting bill shocks
"I woke up to a cloud bill of several million won." Posts like this appear regularly in vibe-coding communities, and they are enough to make anyone's stomach drop. A hobby app has returned a month's salary as an invoice. The scary part is that this is not an accident limited to unusually careless people. If you do not understand the billing model, it can happen to anyone. Billing…
Read article[AI User Guide #8] The AI Agent Era: A Complete Overview
The AI covered in the previous seven installments was ultimately a “ask-and-answer” tool. People drove the conversation: refining questions, providing materials, and verifying answers. But the next step the AI industry has pursued for the past two years is different in kind. AI agents do not just converse—they take on the work itself.…
Read article[Vibe Coder #6] It worked yesterday, not today—why, when I didn’t touch the code?
“I didn’t touch a single line of code, but the app that worked fine yesterday doesn’t work today.” It’s the most common complaint in vibe-coding communities—and one of the most mysterious situations. It feels like a ghost story, but the suspects are actually limited. Investigate them in order, and you can usually catch the culprit before noon…
Read article