サクサク読めて、アプリ限定の機能も多数！
トップへ戻る
GWの過ごし方
cursor.com
We approach building the Cursor agent harness the way we'd approach any ambitious software product. Much of the work is vision-driven, where we start with an opinion about what the ideal agent experience should look like. From there, we form hypotheses about how to get closer to that vision, run experiments to test them, and iterate using quantitative and qualitative signals from evals and real us
Software development is changing, and so is Cursor. In the last year, we moved from manually editing files to working with agents that write most of our code. How we create software will continue to evolve as we enter the third era of software development, where fleets of agents work autonomously to ship improvements. We're building toward this future, but there is a lot of work left to make it ha
Jacob Jackson, Ben Trapani, Nathan Wang & Wanqi Zhu · 7 min read We are observing unprecedented growth in the usefulness and adoption of coding models in the real world. In the face of 10–100x increases in inference volume, we consider the question: how can we take these trillions of tokens and extract from them a training signal to improve the model? We call our approach of using real inference t
Time is a flat circle. When the first version of grep was released in 1973, it was a basic utility for matching regular expressions over text files in a filesystem. Over the years, as developer tools became more advanced, it was gradually superseded by more specialized tools. First, by roughly syntactic indexes such as ctags. Later on, many developers moved to specialized IDEs for specific program
These quality improvements come from our first continued pretraining run, which provides a far stronger base to scale our reinforcement learning. From this base, we train on long-horizon coding tasks through reinforcement learning. Composer 2 is able to solve challenging tasks requiring hundreds of actions. #Try Composer 2 Composer 2 is priced at $0.50/M input and $2.50/M output tokens. There is a
We train Composer for long-horizon tasks through a reinforcement learning process called self-summarization. By making self-summarization part of Composer's training, we can get training signal from trajectories much longer than the model's max context window. This translates into Composer being able to learn to work on challenging coding tasks requiring hundreds of actions. #The limits of compact
Developers are asking coding agents to take on longer, more complex tasks that span multiple files, tools, and steps. As these requests grow in scope, the evals that measure agent performance need to evolve with them. At Cursor, we use a hybrid online-offline eval process to keep our understanding of model quality aligned with what developers actually do. The offline part uses CursorBench, our int
Feb 18, 2026 by Ani Betts, Yash Gaitonde & Alex Haugland in research Coding agents are getting much better at running terminal commands to explore the environment and make changes. Users who auto-approve these commands unlock significantly more powerful agents, but at the cost of increased risk. A mistaken agent can delete databases, ship broken code, or leak secrets. Requiring human approval for
Extend AI agents with specialized capabilities using Agent Skills, an open standard for packaging reusable knowledge and scripts.
Semantic search is one of the biggest drivers of agent performance. In our recent evaluation, it improved response accuracy by 12.5% on average, produced code changes that were more likely to be retained in codebases, and raised overall request satisfaction. To power semantic search, Cursor builds a searchable index of your codebase when you open a project. For small projects, this happens almost
As coding agents became more capable, we found ourselves spending more time on review. To solve this, we built Bugbot, a code review agent that analyzes pull requests for logic bugs, performance issues, and security vulnerabilities before they reach production. By last summer, it was working so well that we decided to release it to users. The process of building Bugbot began with qualitative asses
We've been experimenting with running coding agents autonomously for weeks. Our goal is to understand how far we can push the frontier of agentic coding for projects that typically take human teams months to complete. This post describes what we've learned from running hundreds of concurrent agents on a single project, coordinating their work, and watching them write over a million lines of code a
Coding agents are changing how software gets built. Models can now run for hours, complete ambitious multi-file refactors, and iterate until tests pass. But getting the most out of agents requires understanding how they work and developing new patterns. This guide covers techniques for working with Cursor's agent. Whether you're new to agentic coding or looking to learn how our team uses Cursor, w
Coding agents are quickly changing how software is built. Their rapid improvement comes from both improved agentic models and better context engineering to steer them. Cursor's agent harness, the instructions and tools we provide the model, is optimized individually for every new frontier model we support. However, there are context engineering improvements we can make, such as how we gather conte
We're excited to release a visual editor for the Cursor Browser. It brings together your web app, codebase, and powerful visual editing tools, all in the same window. You can drag elements around, inspect components and props directly, and describe changes while pointing and clicking. Now, interfaces in your product are more immediate and intuitive, closing the gap between design and code. #Rearra
Agent can control a web browser to test applications, visually edit layouts and styles, audit accessibility, convert designs into code, and more.
Composer is our new agent model designed for software engineering intelligence and speed. On our benchmarks, the model achieves frontier coding results with generation speed four times faster than similar models. We achieve these results by training the model to complete real-world software engineering challenges in large codebases. During training, Composer is given access to a set of production
Today, we’re releasing two big updates that make Cursor the best place to work with agents: our first coding model, Composer, and a new interface for working with many agents in parallel. #Introducing Composer Composer is a frontier model that is 4x faster than similarly intelligent models. The model is built for low-latency agentic coding in Cursor, completing most turns in under 30 seconds. Earl
Learn how to use Cursor to build software with AI.
This element contains an interactive demo for sighted users. It's a demonstration of Cursor's CLI showing AI-powered command-line assistance features. The interface is displayed over a scenic painted landscape wallpaper, giving the demo an artistic backdrop.
Our recent pricing changes for individual plans were not communicated clearly, and we take full responsibility. We work hard to build tools you can trust, and these changes hurt that trust. Before we break down how the new Pro pricing works, we want to make sure you know that we will refund any unexpected charges you may have incurred for usage over the past 3 weeks. Timeline: June 16: Original po
You can now work with Cursor Agents on web and mobile. Just like the familiar agent that works alongside you in the IDE, agents on web and mobile can write code, answer complex questions, and scaffold out your work. You can start working them today at cursor.com/agents. Kanban view of Cursor Agents performing coding and research tasks. #What they can do Run tasks while you’re away: Launch bug fixe
Note: We updated this blog on June 30 to improve its clarity. We’re excited to roll out an option to purchase Ultra, a $200 / mo plan with 20x more usage than Pro. While the vast majority of Cursor users are well-served by our Pro plan, this change was highly requested by power users seeking more predictability than usage-based pricing would offer. Ultra is made possible by multi-year partnerships
Bugbot, Background Agent access to everyone, and one-click MCP install Cursor 1.0 is here! This release brings Bugbot for code review, a first look at memories, one-click MCP setup, Jupyter support, and general availability of Background Agent. #Automatic code review with Bugbot Bugbot automatically reviews your PRs and catches potential bugs and issues. When an issue is found, Bugbot leaves a com
Simplified Pricing, Background Agent and Refreshed Inline Edit Introducing unified request-based pricing, Max Mode for all top models, and Background Agent for parallel task execution. Plus, improved context management with @folders support, refreshed Inline Edit with new options, faster file edits, multi-root workspace support, and enhanced chat features including export and duplication. #Simpler
Built to make you extraordinarily productive, Cursor is the best way to code with AI.
This element contains an interactive demo for sighted users. It's a demonstration of Cursor's IDE showing AI-powered coding assistance features. The interface is displayed over a subtle, solid brand background.
This release introduces updates to Bugbot including the ability to self-improve in real time, MCP support, improvements to Bugbot Autofix, and the highest resolution rate to date. #Bugbot Learned Rules Bugbot can now learn from feedback on pull requests and turn those signals into learned rules that improve future reviews. It looks at reactions and replies to Bugbot comments and comments from huma
$20 / mo. Everything in Hobby, plus: ✓ Extended limits on Agenti✓ Access to frontier models✓ MCPs, skills, and hooks✓ Cloud agents
次のページ
このページを最初にブックマークしてみませんか？
『The best way to code with AI』の新着エントリーを見る
j次のブックマーク
k前のブックマーク
lあとで読む
eコメント一覧を開く
oページを開く
