using tools, Cursor Agent have the capability to reason about when it should - read & write code - search your codebase - call MCP servers - run terminal commands an example where it connects to a postgres database, runs a query and writes to file

we've been hard at work improving @cursor_ai Agent, allowing you to delegate more tasks and let it work alongside you agent works just like a human developer, with access to your tools, codebase context, and the ability to take actions here's what Agent can do ↓

with Yolo mode enabled, Agent can autonomously run terminal commands to test and refine code. here's how it can setup a python environment, add a server then run it.

give Agent context through @ commands, just like in Chat. For example, @web enables web searches to pull in relevant documentation and examples, which is particularly valuable when building third-party integrations

agent automatically detects and fixes linting errors using the language server integration. it receives error context and applies the appropriate fixes and keeps on going until the error is resolved. after making changes, it can review them to make sure it lgtm. doing a second pass can often find unexpected changes and bugs

rules can be used to control how Agent behaves. we recently released new project rules, read more about them here

currently available with claude-3.5-sonnet, o3-mini and gpt-4o with more models coming (very) soon!

we have lot's of ideas for new tools and capabilities to offer Agent. let us know if there's any you'd like to see! open a new composer and toggle it on with cmd/ctrl . or click

It’d be cool if someone would make a specialized stack for each Postgres use case, with nice docs, perf comparisons, tutorials, etc - so you don’t have to leave Postgres :)

Right now when I want answers to a question and when I need code change, I have to change from composer to chat which is annoying. This update should improve that workflow by a lot!