Anthropic has made Claude Code's automated code review generally available for Team and Enterprise. Multiple agents read a PR in parallel, cut false positives, and then surface findings in order of severity. Reviews that once took around 20 minutes now run in 6–10, and we break down the pricing, set
AnthropicがClaude Codeの自動コードレビューをTeam／Enterpriseで正式提供。複数エージェントが並行してPRを読み、誤検知を削ってから深刻度順に指摘します。20分前後が6〜10分に短縮され、料金や導入手順、見落としの注意点まで整理しました。
Claude Code v2.1.300 is out. It logs cache writes split by 5-minute/1-hour lifetime, adds a window into subagent logs and a spawn-time rewrite hook. A developer's-eye look at an update that advances observability for unattended, multi-agent operation one more step.
Claude Code v2.1.300が公開。キャッシュ書き込みを5分/1時間の寿命で分けて計上し、サブエージェントのログを覗く窓と起動時の書き換えフックを追加。無人・多エージェント運用の「見える化」を一段進める更新を開発者目線で整理します。
Claude Code v2.1.299's new "idle compaction" automatically compresses conversations over 200K tokens after about 54 minutes of inactivity—before the one-hour prompt cache expires—preemptively avoiding a full re-read on return. MCP permissions were also tightened at the same time.
Claude Code v2.1.299が追加した「アイドル・コンパクション」は、200Kトークン超の会話が約54分放置されると1時間のキャッシュ失効前に自動圧縮し、復帰時の全文読み直しを先回りで防ぎます。MCP権限の狭め直しも同時に。
Linux's Linus Torvalds has published eight page-cache patches he had Claude Code write, pushing small 8-thread reads roughly 3x faster. But he also lays out his own caveats—regressions, untested cases, and where accountability lies.
LinuxのTorvalds氏がClaude Codeに書かせた8本のページキャッシュ改修を公開。8スレッドの小さな読み出しが約3倍に伸びた。ただし回帰や未検証、責任の所在といった但し書きも自ら並べている。
On October 13, Google shipped Gemini 4 Argon to paid plans and integrated its Antigravity development platform inside the macOS Gemini app. We unpack the move to fold everything from building to testing into a single app, along with how to read its pricing and benchmarks.
Googleが10月13日、Gemini 4 Argonを有料プランに配信し、開発基盤AntigravityをmacOS版Geminiアプリの内側へ統合。チャットから実装・テストまでを一つのアプリに畳む動きと、その料金とベンチマークの読み方を整理します。
Vals AI compared GPT-6 Sol and Claude Opus 5.5 as single agents versus teams. Even paying 1.8–5.1× the cost of a single agent, the teams improved scores significantly in only one of four comparisons—putting hard numbers behind the practitioner's instinct that reasoning strength beats head count.
Vals AIがGPT-6 SolとClaude Opus 5.5を1体とチームで比較。チームは1体の1.8〜5.1倍のコストを払っても、成績が有意に伸びたのは4回中1回だけ。「頭数より推論の強さ」という実務の勘どころを数字で示した。
