AI 코딩 도구,
이번엔 뭐가 바뀌었나

• + Rollouts와 Security Review 봇을 출시하여 배포 시 상태를 모니터링하고 보안 취약점을 점검한다 # 보안·권한
  원문

  Rollouts and Security Review: Today we're launching two Cursor bots for the last mile of shipping code. Rollouts watches every change as it deploys and reports its health per environment. Security Review reports exploitable bugs on every pull request. Both are available today on Teams and Enterprise plans.

• + Rollouts가 모든 pull request에서 변경 사항을 모니터링하고 환경별로 상태를 보고한다 # 보안·권한
  원문

  Rollouts: Rollouts attaches a monitor to every pull request and watches the change as it deploys, reporting change health per environment: verified healthy, regression detected, or inconclusive. It's the Cursor version of Firetiger Change Monitors, rebuilt with the Bot Development Kit. Enable it from the dashboard and connect source control, your deploy system, and your telemetry provider. Rollouts starts watching on the next pull request. When a pull request opens, Rollouts reads the diff and the systems it touches, then writes a monitoring plan as a PR comment. The plan lists the risks it identified, the effect the change is meant to have, the signals it will check, and any gaps in instrumentation that would make the change hard to verify. Edit the plan in the PR and Rollouts uses your version. Rollouts wakes on deploy events for the change's commit and runs the plan against your log

• + Security Review가 모든 pull request에서 취약한 버그를 찾아내고 수정 제안을 포함한 리뷰 코멘트를 작성한다 # 보안·권한
  원문

  Security Review: Security Review is available today. It reads every pull request in the context of the codebase and posts one review comment reporting exploitable bugs. Style and quality stay with Bugbot. Security Review comment on a pull request reporting an exploitable bug with a severity and proposed fix Enable it from the dashboard for the repositories you want reviewed. Draft PRs are skipped. Security Review looks for injection across SQL, command, and template surfaces, along with authentication and authorization bypasses, including checks that a refactor stopped running. It also flags secrets and credentials committed to source, SSRF and unvalidated redirects, unsafe deserialization, and dependency changes that introduce known vulnerabilities. It traces where user input enters and what it passes through. Each finding carries a severity, the attack path, and a proposed fix. Dismi

외 1개 항목은 원문에서 볼 수 있습니다.

AI가 만든 요약입니다. 원문과 다르면 오류 제보해 주세요.

• + Claude Opus 5.5(`claude-opus-5-5`)를 추가하며 기본 Opus 모델로 설정한다 # 에이전트
  원문

  [What's changed] Added Claude Opus 5.5 (`claude-opus-5-5`), now the default Opus model — 1M context, $4/$20 per Mtok with $0.20/Mtok cache reads

• + [VSCode] 크롬 확장 프로그램 상태, 설치, 재연결 및 권한 페이지를 포함하는 `/chrome` 명령과 함께 Claude in Chrome 대화 상자를 추가한다 # UI
  원문

  [What's changed] [VSCode] Added a Claude in Chrome dialog (extension status, the install, reconnect and permissions pages, the enabled-by-default setting), opened from the panel menu or by typing `/chrome`

외 112개 항목은 원문에서 볼 수 있습니다.

AI가 만든 요약입니다. 원문과 다르면 오류 제보해 주세요.

• + Claude API, Enterprise, Bedrock, Vertex, Foundry 및 gateways에서 서버측 분류기를 기본값으로 설정하고, 과금되는 대체 방식에 대해 경고를 표시한다 # 에이전트
  원문

  [What's changed] Changed auto mode for Claude API and Enterprise users, and on Bedrock, Vertex, Foundry and gateways, to default to the server-side classifier, which does not charge for classifier overhead (`CLAUDE_CODE_AUTO_MODE_SERVER=0` opts out on Bedrock, Vertex, Foundry and gateways); warns on billed fallback. See https://code.claude.com/docs/en/auto-mode-classifier-billing

외 1개 항목은 원문에서 볼 수 있습니다.

AI가 만든 요약입니다. 원문과 다르면 오류 제보해 주세요.

• + US3, AP1, AP2 및 US1-FED 사이트용 Datadog 프리셋을 추가하고, 새로운 Datadog 연결은 읽기 및 루트 쿼리 API 경로로 제한한다 # MCP·플러그인
  원문

  [What's changed] [Claude Tag] Added Datadog presets for the US3, AP1, AP2 and US1-FED sites; new Datadog connections are now limited to Datadog's read and query API routes

외 95개 항목은 원문에서 볼 수 있습니다.

AI가 만든 요약입니다. 원문과 다르면 오류 제보해 주세요.

• + 조직 제한으로 인해 리뷰가 건너뛰어질 때, 각 원인에 대한 관리자 페이지 링크를 포함하도록 개선한다
  원문

  [What's changed] [Code Review] Improved the check-run card and PR comment when a review is skipped because of an organization limit: each cause now links the admin page that fixes it

외 107개 항목은 원문에서 볼 수 있습니다.

AI가 만든 요약입니다. 원문과 다르면 오류 제보해 주세요.

• + 워크스페이스와 리포지토리 선택기를 통합하고 `sessions.chat.unifiedWorkspacePicker.enabled` 설정을 추가한다 # 에이전트
  원문

  [Agents] Unified workspace and repository picker (Experimental): Setting: sessions.chat.unifiedWorkspacePicker.enabled (Agents window only) Start agent work from one searchable list of local folders, GitHub repositories, Cloud repositories, and remote targets. Remote connection actions remain available from the Remote entry. Selecting Work in Repository uses a cloud-first workflow. A GitHub repository that is not already local is selected immediately with the Cloud harness, without a clone prompt. If you later choose a local harness, VS Code prompts you to clone the repository and preserves the cloud selection if you cancel.

외 11개 항목은 원문에서 볼 수 있습니다.

AI가 만든 요약입니다. 원문과 다르면 오류 제보해 주세요.

• + `claude self-hosted-runner --drain-marker-file <path>` 명령어를 추가한다 # 보안·권한
  원문

  [What's changed] Added `claude self-hosted-runner --drain-marker-file <path>`: when that file exists at a SIGTERM drain, the runner reports its exit to the server as a host drain (telemetry only)

• + Bedrock, Vertex AI, Foundry, LLM 게이트웨이 사용자들을 Claude 데스크톱 앱으로 안내하는 스피너 팁을 추가한다 # UI
  원문

  [What's changed] Added a spinner tip pointing Bedrock, Vertex AI, Foundry and LLM gateway users to the Claude desktop app; the claude.ai desktop app tip now suggests `/desktop`, which offers to download the app

• + Artifact 도구의 게시 오류 메시지를 개선하여, 파일이 없는 경우 먼저 파일에 쓰기를 권장하고 지원되지 않는 파일 형식을 먼저 보고한다 # UI
  원문

  [What's changed] Improved Artifact tool publish errors: a publish with no file now says to write the page to a file first, and an unsupported file type is reported before a missing favicon

• + 자동 모드에서 스킬이나 슬래시 명령의 인라인 `!` 셸 명령이 분류기 대신 기본 모드 권한 규칙을 따르도록 변경한다 # MCP·플러그인
  원문

  [What's changed] Changed auto mode so that a skill's or slash command's inline `!` shell commands follow default-mode permission rules instead of the classifier; a command no rule decides runs as a reviewed tool call

외 92개 항목은 원문에서 볼 수 있습니다.

AI가 만든 요약입니다. 원문과 다르면 오류 제보해 주세요.

• + 플러그인 평가를 위해 `claude plugin eval` 명령어를 추가한다 # MCP·플러그인
  원문

  [What's changed] Added `claude plugin eval`: run a plugin's eval suite against Claude Code and get scored, reproducible results (JSON + HTML report); see `claude plugin eval --help`

• + [VSCode] 에이전트 맵을 추가하여 세션의 서브에이전트 목록과 각 에이전트의 상태를 보여주는 푸터 필을 추가한다 # UI
  원문

  [What's changed] [VSCode] Added an agent map: an "N agents" footer pill opens a map of the session's sub-agents with per-agent cards, Stop agent, and read-only transcripts

• + 관리자 설정에서 GitHub 설치 시 Connect all 또는 Disconnect를 수행하기 전 확인 대화상자를 추가한다 # 보안·권한
  원문

  [What's changed] [Claude Tag] Added a confirmation dialog before Connect all or Disconnect on a GitHub installation in admin settings, to guard against accidental organization-wide changes

외 95개 항목은 원문에서 볼 수 있습니다.

AI가 만든 요약입니다. 원문과 다르면 오류 제보해 주세요.

• + `gatewayInternalNetworks` 관리형 설정을 추가하여 관리자가 조직의 공용 IPv4 블록에서 게이트웨이의 `/login`에 접근할 수 있도록 허용한다 # 보안·권한
  원문

  [What's changed] Added the `gatewayInternalNetworks` managed setting, letting administrators allow `/login` to a Claude apps gateway on their organization's own public IPv4 block

• + `claude plugin` 관련 명령어에 `--json` 플래그를 추가하고 `claude plugin list --json`의 각 행에 `errorDetails`와 `noteDetails`를 추가한다 # MCP·플러그인
  원문

  [What's changed] Added `--json` to `claude plugin install`, `uninstall`, `update`, `enable` and `disable`, and `errorDetails`/`noteDetails` to each row of `claude plugin list --json`

• + Claude가 여러 읽기 전용 조회를 동시에 수행하여 응답 속도를 개선한다 # 성능
  원문

  [What's changed] [Claude Tag] Improved reply speed: Claude now runs several read-only lookups (searching Slack, reading a thread, finding people) at once instead of one after another

외 93개 항목은 원문에서 볼 수 있습니다.

AI가 만든 요약입니다. 원문과 다르면 오류 제보해 주세요.

• + 대규모 작업을 위해 계획을 세우고 서브에이전트에게 작업을 배분하는 Projects 기능을 추가한다 # 에이전트
  원문

  Cursor Projects: Today we're launching Projects in Cursor. Projects lets you take on larger bodies of work, such as a feature, a migration, or a full app. It maintains context over months of work, delegates tasks to thousands of subagents, and performs recurring work without being prompted. Access Projects from the left-hand nav. The coordinator agent in a project doesn't write code itself; it plans the work, delegates it to agents that implement it, and brings the finished work back to you to check. Coordinators create and manage agents on your behalf, running as many in parallel as the work needs.

• + Slack 채널 모니터링이나 정기적 실행 등 외부 신호를 감지하여 자동으로 행동하는 기능을 추가한다 # 에이전트
  원문

  Subscriptions: Tell the coordinator agent to watch a Slack channel, run on a schedule, or follow all your PRs. This way it can take action based on signals it detects, without waiting for your prompt. Connect Slack and point it at a bug-report channel, and it starts delegating each time a bug comes in. Projects are available in beta and rolling out to all users starting today.

외 2개 항목은 원문에서 볼 수 있습니다.

AI가 만든 요약입니다. 원문과 다르면 오류 제보해 주세요.

• + 정기적인 작업을 수행하는 Automations 기능을 추가한다 # 에이전트
  원문

  [Agents] Automations (Preview): Setting: chat.automations.enabled Automations run recurring agent tasks on a schedule, so routine work does not have to be started manually. Start with a template for catching up on changes, triaging issues, or finding bugs, or define your own prompt and schedule. To try automations, enable chat.automations.enabled , open the Agents window, and select Automations in the sidebar. You can run an automation on demand or schedule it to run hourly, daily, or weekly. Learn more about automations in the documentation. Automations is in Preview and is rolling out gradually to all users.

• + 에이전트와 음성으로 대화할 수 있는 Voice Mode를 추가한다 # 에이전트
  원문

  [Chat] Voice Mode (Experimental): Settings: agents.voice.enabled , agents.voice.showTranscript , agents.voice.voice Voice Mode enables you to have a natural, spoken conversation with an agent while it works on your code. Note: Voice Mode is rolling out gradually and might not be available to every eligible account immediately. It requires an eligible individual GitHub Copilot plan and is not available with GitHub Copilot Business or GitHub Copilot Enterprise. Organizations can also turn off Copilot preview features by policy. To try it, enable agents.voice.enabled and select the Voice Mode button in the chat input. While the agent is speaking, start speaking or use the push-to-talk shortcut to interrupt the response and continue the conversation. Voice Mode is aware of your active session and can answer questions about running sessions, the selected model, and attached files. You can also ask

외 9개 항목은 원문에서 볼 수 있습니다.

AI가 만든 요약입니다. 원문과 다르면 오류 제보해 주세요.

• + 풀 요청을 완료하기 위해 에이전트가 피드백을 처리하고 충돌을 해결하며 워크플로우를 재실행하는 Agent merge 기능을 추가한다 # 에이전트
  원문

  [Agents] Agent merge (Preview): Setting: chat.agentMerge.enabled Agent merge helps you take a pull request across the finish line. It asks an agent to address review feedback, fix failed checks and merge conflicts, and rerun workflows. Agent merge repeats this process until the pull request is ready to merge. To try agent merge, enable chat.agentMerge.enabled . You can currently enable agent merge for a session only from the Agents window. Run Enable Agent Merge for Active Session or select the Agent Merge button in the title bar. Learn more about using agent merge.

• + 여러 VS Code 창에서 동일한 에이전트 세션에 연결할 수 있는 Agent host를 도입한다 # 에이전트
  원문

  [Agents] Agent host: The agent host lets you connect to the same agent session from multiple VS Code windows. It runs agent harnesses in a dedicated process based on the Agent Host Protocol (AHP). The agent host's Copilot agent is powered by the Copilot SDK, which aligns the agent's behavior and functionality with the Copilot CLI, the standalone GitHub Copilot app, and other Copilot products. We're actively developing the agent host. The following screenshot shows the Copilot harness selected for an agent host in an editor window: Learn more in the agent host documentation and our new agent host blog post, where we share why we built the agent host, what it enables in VS Code, how the architecture and open protocol work, and workflows you can try yourself. If you have any feedback or requests, please let us know by filing an issue.

• + 기업용 정책을 통해 받아쓰기 데이터의 모델 및 언어 모델 전송 방지 제어를 추가한다 # 보안·권한
  원문

  [Chat] Enterprise controls for dictation data: Administrators can manage the dictation model and language-model transcript cleanup through enterprise policies. The new controls make it possible to require on-device transcription and disable language-model cleanup, so dictation remains available while preventing dictation data from being sent to cloud transcription or a Copilot model. Learn more about enterprise controls for dictation data.

외 12개 항목은 원문에서 볼 수 있습니다.

AI가 만든 요약입니다. 원문과 다르면 오류 제보해 주세요.

• + 로컬 네트워크 내에서 도구 실행을 유지하며 코드베이스와 비밀 정보를 내부 머신에 유지하는 self-hosted machines 기능을 추가한다 # 보안·권한
  원문

  Self-hosted machines: Cursor supports **self-hosted machines**, which let you keep tool execution entirely in your own network. Your codebase, build outputs, and secrets all stay on internal machines running in your infrastructure, while the agent handles tool calls locally. Run on menu showing Remote Machines with self-hosted options lambda-test, cloud-demo, and jacks-desktop

• + 동적 풀 스케줄링을 통해 팀 규모에 맞춰 확장 가능한 self-hosted machines 풀을 관리하고 유휴 머신을 절약하며, Cloud Agents 대시보드에서 상태를 확인한다 # 보안·권한
  원문

  Dynamic pool scheduling: **My Machines** connects a single laptop or VM to your account for personal workflows. **Team pools** are named queues of workers for a team or enterprise. Capacity can grow as requests arrive and shrink when workers disconnect, so your self-hosted machines can scale with demand. Pools are not tied to one repository: name the pool, and any available worker can claim the request. Pools can also hibernate idle machines, then restore within a reconnect window when a follow-up arrives, so you don't keep expensive capacity warm just for the next prompt. Cloud Agents dashboard listing Self-hosted Machines pools with active and idle worker counts

외 2개 항목은 원문에서 볼 수 있습니다.

AI가 만든 요약입니다. 원문과 다르면 오류 제보해 주세요.

• + Android Studio Quail 4를 사용할 수 있게 된다 # UI
  원문

  Android Studio Quail 4 now available

AI가 만든 요약입니다. 원문과 다르면 오류 제보해 주세요.

• + Cloud Agents가 시작할 때 GitHub나 다른 SCM 제공자를 연결할 필요가 없도록 변경한다 # 에이전트
  원문

  Start from scratch, without a repo: Cloud Agents no longer require a connected GitHub or other third-party SCM provider to get started. Prompt from the get-go, then save your work to a Cursor Origin repo.

외 4개 항목은 원문에서 볼 수 있습니다.

AI가 만든 요약입니다. 원문과 다르면 오류 제보해 주세요.

• + 클라우드 에이전트와 Cursor harness를 개선하여 에이전트가 스스로 소프트웨어를 구축하고 배포할 수 있게 한다 # 에이전트
  원문

  Cloud Agents and Cursor Harness Improvements: We're continuing to improve cloud agents and the Cursor harness so always-on agents can operate as a system, building and shipping software on their own without the need for intervention at each loop. With this release, cloud agents can automatically pick up work in response to events, hold a goal until it's met, and stay on course through long-running sessions.

• + 클라우드 에이전트가 이벤트 소스(스레드 또는 대화)를 구독하여 특정 이벤트 발생 시 작동하도록 하는 구독 기능을 추가한다 # MCP·플러그인
  원문

  Subscriptions: Cursor can now monitor your PRs, watch a Slack thread, or run scheduled tasks. Cursor Agent subscribes to an event source (a thread or conversation) and wakes when something happens. Subscriptions are available for cloud agents only, for now. Cloud agents automatically subscribe to PRs they create and drive them to completion, fixing CI and addressing bot comments. In Slack, ask `@cursor check back in an hour and keep going until that feedback is in`.

• + 서브에이전트가 독립적인 가상 머신에서 실행되어 고립된 환경에서 작업을 수행하도록 한다 # 에이전트
  원문

  Subagents on their own machines: Subagents can now run on their own virtual machines. Each gets an isolated copy of the project with clean context in its own cloud environment. Have subagents test the parent agent's changes in fresh environments or swarm independent fixes without collisions. Try `run a swarm of subagents to test my app for bugs, each in its own environment`.

외 3개 항목은 원문에서 볼 수 있습니다.

AI가 만든 요약입니다. 원문과 다르면 오류 제보해 주세요.

• + Cursor가 코드를 호스팅하는 Origin Code Hosting 기능을 베타 버전으로 출시한다 # 에이전트
  원문

  Origin Code Hosting: Cursor can now host your code. Origin begins rolling out today in early beta on all paid plans. We're starting with the essentials, designed for agent scale: repos, pull requests, code browsing, and GitHub sync. Agent-native features ship soon.

• + 모든 리포지토리에서 코드 탐색, 질문, PR 업데이트 및 브랜치 푸시를 수행하는 에이전트 기능을 통합한다 # 에이전트
  원문

  Agents in every repo: Your code, PRs, and agents are now in the same place. Ask Cursor questions about code you're browsing. It can answer, make changes, update PRs, or push a branch.

외 5개 항목은 원문에서 볼 수 있습니다.

AI가 만든 요약입니다. 원문과 다르면 오류 제보해 주세요.

• + 에이전트가 미리 준비된 환경에서 시작할 수 있도록 `Builds` 기능을 도입한다 # 에이전트
  원문

  Cloud Agents Start 3x Faster with Builds: Agents do their best work when they start in a ready environment: repos cloned, dependencies installed, and your install script already run. This release introduces **builds**: ready-to-use copies of your development environment that Cursor prepares in the background. Agents boot into a ready environment instead of setting up from scratch each session. Builds are included with Cloud Agents at no additional cost.

• + 환경 빌드 시스템을 통해 에이전트 시작 속도를 3배 향상시키고 첫 토큰 생성 속도를 10배 향상시킨다 # 성능
  원문

  Faster starts: Cursor runs a new build of your environment regularly. When a build succeeds, it becomes the environment future agents start from. Cursor keeps warm copies ready so the next agent does not wait. Internally, our environments now boot 10x faster, with 3x faster time to first token. Use your install command for anything that can be prepared ahead of time. The start command still runs when you first prompt an agent and should focus on services that need to be fresh in the session.

외 3개 항목은 원문에서 볼 수 있습니다.

AI가 만든 요약입니다. 원문과 다르면 오류 제보해 주세요.

• + AI 에이전트가 실시간 커널을 통해 .ipynb 노트북을 생성, 수정, 실행하며 데이터가 셀 간에 유지되도록 한다 # 에이전트
  원문

  AI agents like Claude Code and Codex can now create, edit, and run .ipynb notebooks against a live kernel, so data persists across cells

• + AI 에이전트가 시스템 환경 대신 프로젝트에 설정된 인터프리터(uv, Poetry, pip, conda)를 대상으로 삼도록 한다 # 에이전트
  원문

  AI agents now target your project's configured interpreter - uv, Poetry, pip, or conda - so packages land in the right environment instead of a system one.

외 9개 항목은 원문에서 볼 수 있습니다.

AI가 만든 요약입니다. 원문과 다르면 오류 제보해 주세요.

• + 여러 VS Code 창에서 동일한 에이전트 세션을 연결할 수 있는 Agent host를 도입한다 # 에이전트
  원문

  [Agents] Agent host: The agent host lets you connect to the same agent session from multiple VS Code windows. It runs agent harnesses such as Copilot, Claude, and Codex in a dedicated process based on the Agent Host Protocol (AHP). The agent host's Copilot agent is powered by the Copilot SDK, which aligns its behavior and functionality with the Copilot CLI, the standalone GitHub Copilot app, and other Copilot products. We're actively developing the agent host and progressively rolling it out to users. The screenshot below shows how to select the Copilot harness on the agent host in the editor window: You can learn more in our VS Code Agent Host documentation. If you have any feedback or requests, please let us know by filing an issue.

외 10개 항목은 원문에서 볼 수 있습니다.

AI가 만든 요약입니다. 원문과 다르면 오류 제보해 주세요.