Bookmarks

Annotate terminal text, review documents and agent replies in Herdr, and send the feedback straight back to the agent. - plannotator/herdr-annotate

Annotate Markdown in the terminal: select, comment, looks-good, delete; send the review to your agent. - plannotator/plannotator-tui

https://t.co/jWyuhe2xUs

A tour of the modern Protobuf workflow, from compatibility and code generation to RPC, validation, and editor tooling, and how each piece builds on the same schema.

Goals, loops, and the discipline of not delegating your judgment

A reference software factory for Claude Code and Codex - addyosmani/factory

A field guide to building a software factory that still has an owner.

I spent a moment to make a project-aware sessions wrapper that works beautifully as a project manager. Super slim solution but works well for me.

Models are commodifying. The harness - skills, extensions, a scriptable app - is where the work and the leverage now live.

はじめに こんにちは、データシステム部MLOpsブロックの木村と、推薦基盤ブロックの上國料です。 ZOZOでは2025年7月より、Claude Codeをはじめとする各種AI開発ツールを利用できる制度を開始しました。 corp.zozo.com 現在ではこの制度のもと、数百名にのぼる社員がClaude Codeを活用し…

YubiKey plugin for age. Contribute to str4d/age-plugin-yubikey development by creating an account on GitHub.

Anthropic's stage-by-stage playbook for the AI-native SDLC: how teams plan, design, build, test, deploy, and maintain software with Claude.

You’ve almost certainly encountered Conventional Commits before. It may have reared its ugly head in the changelog of an open source project you’ve used. It may have been the enforced commit format for an open source project you contributed to. A lot of people swear by it. I swear at it. Even though it is used by a large number of popular open source projects, Conventional Commits is an actively bad standard which encourages focus on the wrong things and fails to deliver on its promises.

Pull requests created by the github-actions[bot] are now able to run your CI/CD workflows with user approval. Requiring approval is a security measure to ensure generated code does not automatically…

Eval しようね。どうも、かわしんです。 昨日 Claude Code に導入したスキルがむしろエージェントの性能を劣化させていたのを発見したのでそれについて実際の実験結果を含めてケーススタディとして紹介します。 コードレビューをしなくなった自分は、AI に作らせたスキルの中身も読んでいません。なんとなく動いていれば…

You don't write AGENTS.md. You train it with gradient descent. - kunchenguid/backpass

And here’s how to properly train it with gradient descent

Native encrypted clipboard sync over SSH for macOS and Linux - standardagents/ssh-clipboard

Peer-to-peer clipboard sync over SSH. Copy anything on one machine, paste on another. No relay, no account, no new ports.

Multi-tier framework for evaluating AI agent skills with quality gates, semantic overlap detection, synthetic evaluation dataset generation, and live agent evaluation that measures how skills affec...

What I've learned from six months of coding with agents—and not writing a line of code by hand.

raindrop.io bookmarklet - an alternative to their browser extension - raindrop-bookmarklet.md

Aperture’s GA release offers built-in tokens, Tailscale MCP endpoints, chat Projects, and fine-grained tool permissions.

AIによる開発支援は「試す」フェーズから「成果を問う」フェーズへ移行しつつあります。 本セッションでは、本イベント参加登録時のアンケート（役職・企業規模・内製開発体制・AIで効果が出ている開発プロセス・導入課題など）の分析結果を発表。さらにファインディが実施するより広範なエンジニア組織調査と比較し、参…

Quality now depends on the constraints you set around your agents.

A fintech company rejected the easy route and homebrewed its own coding agent – and is now a step ahead of coding agents from frontier AI labs. An in-depth look

I have been probably obsessed with two things in the last few years: Nix as a tool to explore innovative ideas that require the capability to rebuild the world and replacing ELF with SQLite as an executable format. You might have noticed that these two ideas are well suited to each other.