AI code editor built on VS Code — agents that edit across your codebase, run in the cloud, review pull requests, and now host your repos.
Cursor is an AI-first code editor forked from VS Code, now built around agents rather than autocomplete alone. Tab still predicts your next edit, and inline chat still answers questions about selected code, but the centre of gravity is the agent: give it a task and it plans, edits across many files with full codebase context, and runs commands. Agents can run locally or in the cloud so work continues after you close the laptop, Bugbot reviews pull requests automatically, and MCP servers, skills, and hooks let you wire in your own tooling.
Cursor Router picks a frontier model per task, and recent releases added an iOS and iPad app, a CLI, and Origin, Cursor's own code hosting with real-time GitHub sync. Because it is built on VS Code, existing extensions, themes, settings, and keybindings carry over.
Pricing model: Freemium
Common questions about Cursor
Cursor's design mode has an Annotate tool for visual prompts. In the Agents Window browser, you can select page elements, draw annotations and double-click text to edit it in place.
Click Annotate or press Cmd + Shift + D. Select several elements with Shift-click, or drag a box around an area. Batch edits keeps your pins together so you can send them as a group.
Annotate requires Cursor 3.24.18 or later. Earlier builds and canvas tabs keep the Design Mode pen, without in-place text editing.
Cursor agents can now control your computer to test the features they build. They can drive desktop apps, including the iOS Simulator, to check their work.
Cursor’s Agent SDK is available for defining an agent as a folder of code, testing it with evals, and deploying it to Cursor or your own infrastructure.
Run npx @cursor/agent-sdk init ./my-bot, then agent-sdk dev to open a local playground. Templates include a security reviewer, a PR autofixer, and a Linear or Jira triager.
The project holds the agent’s instructions, tools, skills, MCP connections and message channels.…
Cursor's cloud agents can now build interactive canvases as part of their work. You can share a finished canvas with your team by sending its link.
The Environments API is also available in beta for creating and managing the saved cloud environments agents run in.
Those environments now support up to 300 repositories. Cursor says large environments also load 5x faster.
Original source
Cursor has opened the Origin App Marketplace to public submissions from teams building developer tools. It also has over 20 integrations across deployments, CI, code review, security, monitoring, and project management, including Cloudflare, Linear, and Sentry.
To submit an app, build it privately in your namespace, then make it public and submit it from its page. You'll need a slug, suggested tags, and a demo video showing installation, uninstallation, and the core product flows.…
You can now check on local Cursor agents and send them messages from the Cursor iOS app.
Sign in on iOS, select a computer on your account, then approve the pairing request in Cursor's desktop app. Tap an agent to see its activity or message it.
Agents keep running on your computer, which must stay on and online. Cloud agents aren't required. To use “Keep this computer awake” in the desktop Remote Control settings, the computer needs to be plugged in with its lid open.…
Cursor has launched Rollouts and Security Review for Teams and Enterprise plans.
Rollouts monitors changes as they deploy, checking logs, metrics, and traces separately for each environment. Connect source control, your deploy system, and your telemetry provider to get started. It posts a monitoring plan on each PR that you can edit. When it detects a regression, it can open a revert PR for review or hand the finding to a cloud agent, depending on configuration.…
Claude Opus 5.5 is now available in Cursor. Cursor says it leads CursorBench at 57.8% in Max mode and costs 40% less per task than Opus 5.
Read the original announcement →
Cursor Projects gives larger jobs, such as a feature, migration, or full app, a persistent coordinator agent. It plans the work, delegates coding to other agents, and brings the results back for you to check.
Each Project runs on its own cloud computer and keeps going when you close your laptop. Shared files carry context across its agents, and the coordinator can start a local agent when testing needs your machine.…
Cursor's cloud agents can now execute tools on your own infrastructure, giving them access to internal services or specialized hardware on your network.
Planning and inference stay in Cursor's cloud. Worker pools handle tool execution, scale automatically with demand and can serve several repositories.
Read the original announcement →
Other products to explore and follow.
The people behind Cursor
Build, refine, and publish apps, websites, and other projects by describing what you want to Replit Agent.
Cloud-based AI coding agent by OpenAI — assign tasks and Codex writes, tests, and ships code in a sandboxed environment.
Anthropic's agentic coding tool — Claude in your terminal, IDE, desktop app and browser, reading and editing your codebase, running commands, opening pull requests and working on tasks in the cloud.
Spotify's vendor-neutral agentic development environment — run and manage agent sessions across Claude Code, Gemini CLI, and Codex from one place.
Moonshot AI assistant for coding and agentic work. 3D Make connects to tools like Blender and CAD tools.
Agentic IDE, formerly Superset, for running coding agents such as Claude Code, Codex, and OpenCode in one workspace.
