話題 / Topic
Claude Code v2.1.287 introduces "Mods," running JS/TS handlers inside the core process so they can replace everything from tool calls to screen rendering. They're on by default, but they run with the same privileges as Claude Code itself — making it essential to vet your distribution sources.
Claude Code v2.1.287 が「Mods」を導入。JS/TSのハンドラを本体内部で走らせ、ツール呼び出しや画面描画まで差し替え可能に。既定で有効だが権限はClaude Code本体と同じで、信頼できる配布元の見極めが要る。
On September 28, Anthropic added build-eval and hillclimb to Claude Code's claude-api skill. They semi-automate eval creation and self-improvement, with built-in safeguards against overfitting, replacing "tuning by instinct" with tuning by the numbers.
Anthropicが9月28日、Claude Codeのclaude-apiスキルにbuild-evalとhillclimbを追加。評価づくりと自己改良を半自動化し、過学習を自分で疑う仕組みで「勘のチューニング」を数値に置き換える。
Claude Code v2.1.286 (released September 30, 2026) changes behavior so that when the default model is rejected by the API, it retries once with the previous model of the same tier. It closes the "every-turn failure" in unattended runs, and also refines the approval counter display and how remote-con
Claude Code v2.1.286（2026年9月30日公開）は、既定モデルがAPIに拒否されても同じ格の前のモデルで一度だけ再試行するよう変更。無人運用の「全ターン失敗」を塞ぎ、承認の件数表示やリモート操作の権限反映も整えた。
Claude Code v2.1.285 adds allowedProviders to limit which API endpoints may be used, CLAUDE_CODE_DISABLE_WEB_FETCH to turn off WebFetch, and a sandbox that lower-level settings can't loosen. In the age of unattended operation, the controls for administrators to tighten the agent's "outbound mouth" a
Claude Code v2.1.285は、使ってよいAPI接続先を絞るallowedProviders、WebFetchを無効化するCLAUDE_CODE_DISABLE_WEB_FETCH、そして下位設定で緩められないサンドボックスを追加。無人運用時代の“外向きの口”を管理者が締める設定がそろった。
Claude Code's default model was updated twice in one week. v2.1.284 makes Sonnet 5.5 the standard, while mechanisms like deniedModels and a dollar-denominated /usage put 'which models, and up to how much' in administrators' hands.
Claude Code の既定モデルが1週間で二度更新。v2.1.284がSonnet 5.5を標準に据え、deniedModelsや金額表示の/usageなど『どのモデルを・いくらまで使うか』を管理者が握る仕組みが同時にそろった。
Claude Code v2.1.257, released September 1, 2026, swaps its default model to Fable 5.1 with its one-million-token context. It also adds guardrails to auto mode that stop credential retrieval and out-of-scope reads from slipping through. Here's a rundown of the changes that matter to developers.
2026年9月1日公開のClaude Code v2.1.257が、既定モデルを100万トークン文脈のFable 5.1へ差し替え。自動モードには資格情報取得や範囲外読み取りを素通しさせない歯止めを追加した。開発者に効く変更点を整理する。
