Changelog

Slash commands, summarization and improved Agent terminal

Custom slash commands

You can now create reusable prompts and quickly share them with your team. Commands are stored in .cursor/commands/[command].md. Run them by typing / in the Agent input and selecting the command from the dropdown menu.

We've been using them for running linters, fixing compile errors, and creating PRs with detailed descriptions and conventional commits.

Summarization triggers

Cursor automatically summarizes long conversation for you when reaching the context window limit. You can now summarize context on-demand with the /summarize slash command. This can be useful when you don't want to create a new chat, but want to free up space in the context window.

MCP Resources support

We've added support for MCP Resources. Resources allow servers to share data that provides context to language models, such as files, database schemas, or application-specific information.

Additionally, interpolated variables are now supported for MCP. This enables using environment variables in strings when defining configuration for MCP servers.

Improved terminal for Agent

When Agent decides to create a terminal to run shell commands, we've dramatically improved the stability and reliability of the environment.

This solves known issues around terminal commands hanging and not properly exiting when completing tasks, as well as improving the SSH experience.

We've also polished the terminal UI, made it faster to run, and added OS notifications when shell commands require user approval.

Linear integration, improved Agent terminal, and OS notifications

Run Agents in Linear

You can now start Background Agents directly from Linear. Delegate tasks to Cursor without leaving your issue. We've written a longer blog post with more details.

Improved review flow in Agent terminal

The terminal now opens on the left with a clear backdrop and border animation to highlight when it’s blocking. Rejecting auto-focuses the input so you can respond immediately.

OS Notifications from Agent

Get native OS notifications when an agent run finishes or when input is required, for example approving a command that is not allowlisted. Enable from Settings.

Support for MCP elicitation

Cursor now supports MCP elicitation, a new feature in the MCP spec that allows servers to request structured input from users, such as a user preference or configuration choice. Requests are defined with JSON schemas, giving servers validated responses while ensuring users stay in control of what they share.

Improved Agent tools, steerability, and usage visibility

More Agent steerability

Sending messages while Cursor is working now does a better job of steering the agent. Messages will run at the next ideal time, usually after a tool call, versus waiting until the generation completes. ⌥+Enter (Alt+Enter on Windows) will queue messages like before and ⌘+Enter (Ctrl+Enter on Windows) will interrupt the agent to send the message immediately. The default behavior can be changed in Cursor Settings -> Chat -> Queue messages.

Improved Agent tools

Agents are significantly improved when operating across large codebases. Context selection is more relevant, token use is more efficient, and edit quality is higher:

• Read file: Now reads full files when appropriate and the 2MB cap is removed.
• List: Can now explore full directory trees in one call, with metadata like file counts and types.
• Grep: Improved matching with less noise.
• Codebase Search: Better ranking and indexing, leading to more relevant context.
• Web Search: Tuned for smaller, more relevant responses using a lightweight native model.

Separate models per Agent

You can now assign a different model to each Agent in separate tabs and editors. Forking an Agent preserves the selected model.

Usage and pricing visibility

Usage stats are now visible in Chat. You can see your usage either all time, or when you've exceeded 50% of your plan's quota.

This summary can be toggled in Settings. Team and Enterprise users will have access to the usage summary in September.

Compact chat mode

Compact mode hides tool icons, collapses diffs by default, and auto-hides the input when idle. This can be useful for long sessions with many tool calls, terminals, and diffs. You can toggle compact mode on from settings.

GitHub support for Background Agents

Agents can now be used directly inside GitHub pull requests. Tag @Cursor and the Agent will read the prompt, apply the fix, and push a commit.

Background Agent will keep you updated on status with a comment and todos

Sidebar for all Agents

You can now inspect and manage all Agents from the left sidebar, both foreground and background. Peek into remote machines to see Background Agent working by clicking on one in the sidebar.

Faster Background Agent startup

We’ve put a lot of effort into improving Background Agents and they are now twice as fast to start.

Shared terminal with Agent, context usage in chat, and faster edits

Share terminal with Agent

Agents can now use your native terminal. A new terminal will be created when needed and runs in the background if one isn’t already open. Click Focus to bring it up front where you can see Agent commands and also take over.

View context usage in Chat

At the end of a conversation you can now see how much of the context window is used.

Faster edits

Agent edits are now faster by lazy loading linter errors. Search & Replace edit latency has been reduced by 25% and Apply edits by almost 11%.

Agent Planning, Better Context & Faster Tab

Agent To-dos

Agents now plan ahead with structured to-do lists, making long-horizon tasks easier to understand and track.

The agent breaks down longer tasks with dependencies, visible to you in chat and streamed into Slack when relevant. It can update this list as work progresses, keeping context fresh and interactions predictable.

To-dos are also visible in Slack if integration is setup!

Queued messages

You can now queue follow-up messages for Agent once it's done with current task. Just type your instructions and send. Once in queue, you can reorder tasks and start executing without waiting.

Memories (now GA)

Memories is now GA. Since 1.0, we've improved memory generation quality, added in-editor UI polish, and introduced user approvals for background-generated memories to preserve trust.

PR indexing & search

Cursor now indexes and summarizes PRs much like it does files. You can search old PRs semantically or explicitly fetch a PR, issue, commit, or branch into context.

This includes associated GitHub comments, BugBot reviews, and Slack agent support—making postmortem analysis or incident tracing much faster.

Improved embeddings for semantic search

Codebase search is now much more accurate with our new embedding model. We've also re-tuned prompts to yield cleaner, more focused results.

Faster Tab

Tab completions are now ~100ms faster, and TTFT has been reduced by 30%. We made this possible by restructuring our memory management system and optimizing data transfer pathways.

Let Agent resolve merge conflicts

When merge conflicts occur, Agent can now attempt to resolve them for you. Click Resolve in Chat and relevant context will be added to resolve the conflict.

Background Agent improvements

Several improvements to Background Agents make them more predictable and resilient:

• PRs follow your team's template
• Changes to the agent branch are auto-pulled in
• Conflicts (like rebases) are now surfaced as actionable follow-ups
• You can commit directly from the sidebar
• Slack and web deeplinks open the associated repo, even if you don't have it open