Bookmarks

TL;DR * Devinは長時間タスクの完走能力が他のエージェントより優れています。その分コストも高いです。 * Claude Code Actionはタスク実行速度が最も速く、成功率も高いです。コストパフォーマンスも高いです。 * その他のエージェントは内部セッションタイムアウトがあり、タスクを中断します。長時間タスクには向きません。 最終結果 エージェント名 完了問題数／実行時間 コスト 1問あたり 正解数／正解率 結果 🏅Devin 98問／216分 $36 $0.37 92問／91.1% 長時間タスク完遂能力抜群、コスト高 🥈Claude Code Action 92問／42分 $7.89 $0.09 65問／64.4% 最速・高コスパ 🥉GitHub Copilot Coding Agent

津波による浸水の深さが1mになると、想定死亡率は100％となります。津波が予想される事態となれば、一刻も早く避難し、巻き込まれることを防ぐのが最も重要です。

We're launching a secure, independent authenticator app available on desktop and mobile, to help you secure your online accounts.

You deserve some always-on gadgets—and an easier way to access them.

Building a fully functional, code-editing agent in less than 400 lines.

Discover how Anthropic approaches the development of reliable AI agents. Learn about our research on agent capabilities, safety considerations, and technical framework for building trustworthy AI.

Anthropic released two major new features for their consumer-facing Claude apps in the past couple of days. Sadly, they don’t do a very good job of updating the release notes …

Use Claude Code as the foundation for coding infrastructure, allowing you to decide how to interact with the model while enjoying updates from Anthropic. - musistudio/claude-code-router

TypeScript language bindings for the Pkl programming language - pkl-community/pkl-typescript

QakeNotch makes your MacBook Notch useful with a quake notch terminal and other cute stuff.

like gofmt, but for jq. Contribute to noperator/jqfmt development by creating an account on GitHub.

Look up macOS system binaries information including executables, frameworks, and application bundles.

Declarative secrets, every environment, any provider. - cachix/secretspec

モダン並行プログラミングパターンとベストプラクティス

Description will go into a meta tag in head /

A powerful coding agent toolkit providing semantic retrieval and editing capabilities (MCP server & Agno integration) - oraios/serena

Fast caching software with a focus on low latency and cpu efficiency. - tidwall/pogocache

AWS の新機能についてさらに詳しく知るには、 Amazon EventBridge がオブザーバビリティ向上のためにログ記録のサポートを開始

Transform ideas into high-quality apps with an AI-native platform that takes you from concept to deployment in moments—without sacrificing speed or control.

はじめに こんにちは。SREチームの侘美です。 弊社ではLLM（大規模言語モデル）を活用したコーディング、特にDevinやClaude Codeのようなエージェント型ツールを積極活用する方針を打ち立て、各種ツールを利用できる環境を整備しています。 また、習熟やノウハウの獲得のため各チームで一定期間AIエージェント縛りで…

Explore data and build apps seamlessly with marimo, a next-generation Python notebook.

Git/Githubの実践的な使い方やオープンソースでコラボレーションする作法を学ぶための日本語のチュートリアルです！

Securing open-source package ecosystems by originating, validating, and augmenting build attestations. - google/oss-rebuild

Stop treating AI evals as a silo. Run them like unit tests, via vitest-evals, JUnit XML, and Codecov, to keep quality tight and releases safe.

A vitest extension for running evals. Contribute to getsentry/vitest-evals development by creating an account on GitHub.

The dispatch-method is a way multiple Stimulus controllers can communicate with each, and trigger user interface interactions

Lightweight, performant interactive network connection monitor with friendly service names - taylorwilsdon/netshow

A terminal user interface (TUI) tool for trimming OpenAPI specifications down to size. Apisnip allows you to interactively select which endpoints to keep in your API specification, making it easy t...