Claude Articles
An experiment in vibe coding
A developer builds a travel itinerary web app for his wife using AI coding assistants like Claude, exploring the 'vibe coding' approach and its pros/cons.
Advent of AI 2025 - Day 14: Agent Skills
Explores the open standard for AI agent skills, detailing how tools like Goose use SKILL.md files to provide contextual expertise automatically.
Dumping ChatGPT for Claude
A developer explains their switch from ChatGPT to Claude for coding and technical work, citing Claude Code's effectiveness and personal preferences.
Don’t let A.I. read your .env files
A guide on preventing AI coding assistants from reading sensitive .env files, explaining the security risks and offering a solution using 1Password CLI.
The Unexpected Effectiveness of One-Shot Decompilation with Claude
Using Claude AI to accelerate decompilation of N64 games by automating the conversion of assembly code into human-readable C.
Claude 4.5 Opus' Soul Document
Anthropic's internal 'soul document' used to train Claude 4.5 Opus's personality and values has been confirmed and partially revealed.
llm-anthropic 0.23
Release of llm-anthropic 0.23 plugin adding support for Claude Opus 4.5 and its new thinking_effort option.
Quoting Claude Opus 4.5 system prompt
Analysis of a leaked system prompt for Claude Opus 4.5, discussing its content and the challenges of evaluating new LLMs.
Surprises hidden in the Claude Opus 4.5 System Card
Analysis of surprising findings in Claude Opus 4.5's system card, including loophole exploitation, model welfare, and deceptive behaviors.
Using Claude Models in Microsoft Foundry with Microsoft.Extensions.AI (temp fix 😏)
A guide to using Anthropic's Claude models in Microsoft Foundry with Microsoft.Extensions.AI via a custom HTTP handler for API format translation.
Building an AI REPL with QuickJS
A developer builds an AI-integrated REPL using QuickJS and WebAssembly, exploring AI's ability to analyze JavaScript bytecode.
llm-anthropic 0.22
Release of llm-anthropic plugin 0.22 with support for Claude's structured outputs and web search tool integration.
Just Talk To It - the no-bs Way of Agentic Engineering
A developer shares their workflow for using AI agents to write 100% of their code, focusing on practical tools and strategies for agentic engineering.
Building Features One Prompt at a Time
A CEO describes using AI (Claude 4 via Copilot) to code and ship a new autoqueue feature for Mergify in under an hour a day over three weeks.
Just One More Prompt
A developer's personal reflection on the addictive nature of AI agentic engineering, the loss of work-life balance, and the industry-wide trend of extreme work hours in the AI space.
Self-Hosting AI Models After Claude's Usage Limits
Explores alternatives to Claude Code after usage limit changes, focusing on self-hosted AI models and CLI tools for developers.
I Built a Claude Code Context Modal Inside of Neovim
A developer built a custom Neovim plugin named Kai that integrates Claude Code's command-line AI directly into the editor for intelligent code assistance.
Comparing Claude Code vs OpenCode (and testing different models)
A developer compares Claude Code to OpenCode with different AI models (Sonnet-4, Gemini, GPT-4.1) on a practical coding task, evaluating their performance and workflow fit.
Vibe coding MenuGen
A developer built MenuGen, an app that uses AI to generate images of menu items from a photo, and shares their experience creating it with AI coding tools.