#coding-agents

Properties of a Good Multi-Turn Agent Harness

A multi-turn agent harness can be understood as the system that maintains a reliable interaction loop between an agent and the environment…

Behind Every AI Coding Assistant Is a Hidden System — Here’s How It Works

AI coding assistants have changed the way we write software.

From Coding to Orchestration: Systems Thinking in the Age of Agentic Software Engineering

Why AI Coding Agents Make Systems Thinking the New Core Skill in Software Engineering

GLM-5.2: Open-Source 1M Context AI Outperforms Giants

Introduction

What I Learned Evaluating Cline and Cursor in Production

The lesson that stuck with me came from two failures. A developer points Cline at a feature branch overnight, the autonomous loop runs for…

How to integrated GLM 5.2 in Visual Studio Code

To use z.ai (Zhipu AI / GLM models) as a coding agent in Visual Studio Code, you have two main options.

Analysis of Anthropic’s 400,000 Recent Claude Code Sessions: What Matters More Than AI Writing Code

Stop Prompting Coding Agents. Start Designing Loops Instead.

The biggest shift in AI isn’t better prompts. It’s building systems that can improve themselves without waiting for your next instruction.

The Frankenstein Codebase — Part 1

The day when your once-cohesive project starts to feel like it was written by four different engineering teams. The Multi-Model Toxicity…

Swival 1.0.27 is a big release

Swival 1.0.27 is out. And it’s a big release.

How to build a coding agent — Part 1.0

In this tutorial series, we build a coding agent that you can assign tasks on GitHub. In this part 1.0, we discuss:

17 Open Source Projects You Need

MCP tools and coding agents

Software Engineering as the Interface Between Humans and AI

After several months of agentic development, I found that generated code was abundant while confidence in the generated system stayed…

How to Create an Environment to Work with Many Coding Agents

In this guide, you will learn how to work seamlessly with Claude Code, Codex, Cursor, etc.

How We Automated a 3-Month Refactor in 3 Weeks Using AST Parsing and a Coding Agent [Part-1]

A technical deep-dive into building an AI-powered pipeline that migrated 170+ files of raw Django ORM calls to a clean Data Access Layer —…

ProcCtrlBench: Evaluating Process-Level Defects and Control Preservation in LLM Coding Agents

Paper Title: ProcCtrlBench: Evaluating Process-Level Defects and Control Preservation in LLM Coding Agents

I Gave an AI the Ability to Write and Run Its Own Code.

The most dangerous program is one that runs without anyone watching.” — every senior engineer who has debugged production at 2 a.m.

I Found the Best Claude Code GUI (That Works with Codex & Gemini Too)

If you want to keep coding and forget the AI coding agent wars, you need to try one of the best AI agent GUIs out there.

Embedding the Everruns Runtime: A Local Coding Agent in One Binary

Most of what we write about Everruns describes the server: durable execution, persistent state, multi-day runs that survive crashes. That…

How to Safely Run Coding Agents

Coding agents such as Claude Code and Codex have provided me the biggest efficiency boost I’ve ever experienced while programming, way more…

The Industrialization of Agentic Software Engineering and AI Pricing 2026

The software engineering landscape in 2026 has shifted from simple autocomplete to autonomous agentic software engineering, where agents…

Cline : Open Source Agentic Ecosystem Across SDK IDE CLI

Introduction

AI Agent Instruction Ladder: 8 Levels, 3 Channels, 2 Cuts

A few months ago I drew a maturity ladder for CLAUDE.md files — does the file exist, are constraints explicit, do skills load on demand…

[Nex-AGI / Fudan University] — Agentic Harness Engineering: Observability-Driven Evolution of…

1. Executive Summary & Background

The 12MB AI Coding Agent: Why I’m Ditching Bloated Dev Tools for ZeroStack

If you’ve experimented with modern AI coding agents, you’re probably familiar with the trade-off. They are incredibly capable, but they are…

Three Patterns Where Agent-Generated Code Quietly Fails

And what that means for your skill stack as a senior engineer

Coding Agent Index 2026 — Benchmarking Full Agent Stacks (Model + Harness)

Coding Agent Index is the first independent benchmark of full agent stacks. What it measures, what it doesn’t cover, and how to evaluate…

Generating Editable draw.io Diagrams in VS Code with MCP

Technical diagrams can be generated directly inside VS Code using the draw.io MCP together with GitHub Copilot Chat. This setup supports…

The Four-Loop Pattern: Running Autonomous Coding Agents Without Drift

How to keep AI engineering disciplined when the agents work faster than you can review them — including when the author of the discipline…

Autonomous Coding Agents — Which one to choose?

Okay, let’s be real. Every week, there’s a new autonomous coding agent with a different price. And everyone online swears theirs is the…

KAIST & NYU — Memory Transfer Learning: 코딩 에이전트를 위한 교차 도메인 지식 전이 아키텍처 분석

1. Executive Summary & Background

Anthropic’s Managed Agents: I Read the Fine Print So You Don’t Have To

Anthropic launched Claude Managed Agents on April 8, 2026. The internet called it a game-changer. Here’s what’s real, what’s still in…

[arXiv] — Coding Agents are Effective Long-Context Processors

1. 배경 및 개요

[Stanford / MIT] — Meta-Harness: 모델 하네스의 종단간 최적화를 통한 LLM 시스템 성능 극대화

1. 배경 및 개요

Ideas Are Cheap No More or Lessons Learned After A Year of Vibe Coding

As many tech folks, I decided to onboard myself into vibe coding when it became a mainstream and today I’m ready to share my learnings.

Coding Agents. The Insider Threat You Installed Yourself

Stop Running AI Coding Assistants Blindly

Make your Claude Code make nice diagrams — Introducing trickery

Couple of days ago I wanted to generate nicer diagrams for one of my projects, my first attempt was to use mermaid, which worked, but not…

Making Your Repository AI-Ready

I’ve been using AI coding agents for a while now. I mostly use GitHub Copilot, Cursor and Claude Code, but the same would apply to any…

When Code Clashes: A Tale of Two AIs Solving a Network Puzzle

It was a frustrating morning. My internet connection decided to misbehave, and my home development server — along with the virtual…

Top 5 Best AI Coding Agents

When you’re running a lean startup like Kommunicate, you build many tools yourself. For example, I have a small AI agent that analyses all…

Gemini CLI, yet another terminal-based AI-assisted

Reimplementing a Zig CLI tool with GraalVM and Java using Gemini CLI

OpenAI Codex CLI: A Developer’s Guide

Codex

Darwin Gödel Machine

Below is beginner-friendly vibed blog interpreting Darwin Gödel Machine: Open-Ended Evolution of Self-Improving Agents paper.

Amdahl’s Law and the Myth of 10x Developers in the AI Age

In the rapidly evolving landscape of software development, we’re witnessing a surge in AI coding assistants and the eternal pursuit of the…