Bookmarks

Codex の Record & Replay は macOS 上でのユーザーの操作を実演することで再利用可能なスキルに変換する機能です。例えば経費精算の提出や勤怠アプリへの打刻や工数入力、定期的なレポートの作成などをスキルとして記録し、煩雑な定型業務を AI に任せることが期待できます。この記事では、Record & Replay を実際に試してみた様子を紹介します。

OSS開発者は今何をするべきか？ソフトウェアサプライチェーン侵害対策を考える June 23, 2026

Sakana AI Console Provider in ~/.pi/agent/models.json - models.json

Learn about the seven methods for instructing Claude's behavior to understand the context cost and authority of each.

The post examines Ponytail, a popular AI coding “skill”, and argues that its benchmarked benefits appear to come largely from encouraging terse, YAGNI-style responses rather than from any deeper engineering value. By showing that a simple prompt can match or beat Ponytail on its own benchmark, it makes a broader case for treating prompt-based tools with scepticism unless their claims are backed by robust evaluation.

Makes your AI agent think like the laziest senior dev in the room. The best code is the code you never wrote. - DietrichGebert/ponytail

SSH port forwarding explained in a clean and visual way. How to use local and remote port forwarding. What sshd settings may need to be adjusted. How to memorize the right flags.

Vercel が新しい AI エージェントフレームワーク eve を発表しました。Next.js の設計思想に基づいて構築された eve は、AI エージェントの開発に必要な機能がすべて揃ったフレームワークです。この記事では、eve を使って簡単なエージェントを作成し、実行する方法を紹介します。

Like Next.js for web apps, but for agents. Markdown for instructions and skills, TypeScript for tools. Durable by default.

Sudoless Apple Silicon system monitor (native SwiftUI GUI) with ANE / Media Engine / memory-bandwidth tracking - kennss/SiliconScope

Docker-compatible REST API on top of Apple container - socktainer/socktainer

AA-Briefcase is our new benchmark for testing models on long-horizon knowledge work tasks in complex projects built by industry experts. Models are evaluated on multi-week projects, each with many

CTO の川口(id:dmnlk)です。 最近、社内で「とりあえず HTML にして置いておくね」という会話を当たり前のように耳にするようになりました。きっかけは、社内 HTML をホストするだけのささやかな環境をひとつ用意したことです。たったそれだけのことなのに、気づけばエンジニア以外のメンバーまで使い始め、社内の情…

The all-in-one, open-source backend platform for agentic coding. InsForge gives your coding agent database, auth, storage, compute, hosting, and AI gateway to ship full-stack apps end-to-end. - Ins...

Model gateway, compute, deployment, database, auth, and more. Every service built for AI coding agents to operate end to end through CLI and skills.

Shadow any website for offline viewing, with the JavaScript stripped out - tamnd/kage

In the previous article we watched the runtime rebuild an entire stack trace out of metadata the compiler and linker had frozen into the binary at build time. I told you at the end that reflect works on exactly the same trick — metadata baked into the binary, only pointed at your data instead of your call stack. Today we’re going to cash that promise in. Let’s start with a program that, the first time you see it, feels like it shouldn’t be possible:

Large DELETEs add work instead of reclaiming it. Structure your database so deletion becomes DROP TABLE or TRUNCATE.

Fear-free migrations for PostgreSQL. Contribute to gocardless/nandi development by creating an account on GitHub.

MDN's MCP server brings MDN's documentation and browser compatibility data directly into your editor or IDE, giving your LLM or coding agent access to accurate, up-to-date web platform information.

📌 Overview Ruby on Rails has helped make it possible to scale out the database layer, meeting the demands of millions of Aura Frames customers enjoying their digital photo frames. In late 2025, the team added additional primary databases to expand capacity for peak write and read load ahead of Christmas Day, the busiest day of the year for the company. Rails manages queries and schema changes for each primary database within the same codebase, and now with the additional capacity of many primary databases. With 8 primary databases in total, each server instance can be vertically scaled ahead of peak load. When load returns to normal levels, instances are scaled down for cost savings. The team leveraged native support for Multiple Databases and the disable_joins: true feature in Active Record, the ORM for Ruby on Rails. The disable_joins feature replaces SQL joins, issuing multiple SELECT statements to combine data in the application from different databases. This post looks back at the technical details of that plan, as well as a variety of additional data layer scaling tactics, that culminated in a successful Christmas 2025 season, with peak U.S. and Canadian Apple App Store and Google Play Store rankings of #1.

This post was originally given as a talk at Rubycon IT 2026. The slides are also available. It’s been a while since I first talked about rv, a Ruby manager for the future. I’d like give an update on what we’ve done since then, but I’m going to recap some of that earlier post first to give context for the updates. If you still remember what I said back then, you can jump to the new stuff right away. Either way, I’m excited to update you about the work that we’ve been doing, and show exactly how far we’ve gotten.

Tried to use env vars to set versions, didn't work. Wasted a couple hours testing several different configs.

React Router v8 is available now.

In this essay, Carson Gross argues that as AI makes code cheap to produce, understanding code becomes the expensive and scarce resource. He warns of the complexity that LLM code can generate and proposes the subtractive, constraining engineer as the discipline needed to keep systems comprehensible & stable.

Install pnpm and a JavaScript runtime (Node.js, Bun, or Deno) in one step

How numpy-ts went from 15x slower than NumPy to performance parity