Cursor
AI coding agent
5.0•894 reviews•50K followers
AI coding agent
5.0•894 reviews•50K followers
Cursor is the best way to build ambitious software.
This is the 10th launch from Cursor. View more
Cursor for iOS
Launched this week
Cursor is now available as a native iOS app in public beta, so you can build from anywhere.
Until now, developers have worked around the limits of their local machines, keeping laptops half-open and caffeinated everywhere they go.
With Cursor for iOS, you can launch always-on agents in the cloud, or control agents running on your computer from your phone. Kick them off when ideas strike, get notified when work is ready for review, and merge PRs on the go.
Launch Team
Codex by OpenAI
Claude Code
Antigravity IDE for Vibe Coding
The always on cloud agents caught my attention . I'd love to hear how they're performing for people using them on real projects using them on real projects.
Cursor
@bernard_lewis Internally, more than 40% of all our PRs (not just from mobile) come from cloud agents, and growing. Hope you give it a try and let us know what you think!
Great launch 👏 I’m curious how many developers will end up using the phone for quick reviews versus actually starting new coding tasks.
Cursor
@rahul_manjhi1 we saw developers use the mobile app in both of those ways during the private beta! Always-on cloud agents make it especially easy to kick off a new task without needing to pair with your computer beforehand.
This is an interesting shift . Mobile won’t replace the desktop IDE , but being able to review code, manage agents , and keep projects moving on the go could be real productivity boost .
Cursor
@khushi_ranwa we've found it to be a huge productivity boost! Not a replacement for all desktop work, but our early beta testers were surprised by how much development they're able to do entirely from the mobile app.
Cursor
@ayush_pal8 Cursor for iOS requires a paid plan. Hope you try it out!
I switched to Kilo after being burned by Cursor's predatory billing practices. Cursor charged me $313 in "Premium model overage" fees on top of my $200/month Ultra plan — charges they never disclosed upfront. Even their own AI Support Agent admitted the marketing was misleading. They refused to refund. I had to dispute the charges with my bank.
Kilo has been the exact opposite experience.
I brought my own DeepSeek API key — and it just worked. No vendor lock-in. No hidden fees. No "oh by the way, that model costs extra" surprises. I can use DeepSeek V4 Pro alongside other models at the actual provider rates, without any markup.
What I love about Kilo:
Honest pricing — you pay for what you use, and you see exactly what it costs
Model freedom — switch between models mid-session without rebuilding your workflow
DeepSeek support — finally able to use it at its full potential without restrictions
Open source — you can audit everything that runs on your machine
Cursor turned a loyal, paying customer into an activist.
Kilo turned me back into a productive developer.
If you care about transparency, control, and not being nickel-and-dimed by your AI tools, Kilo is the answer.
one suggestion from me would be to add a quick activity timeline showing everything an agent has done while I was away. I think that would make it even easier to catch up before approving changes.
Cursor
@scarlett_massey love this idea! Thanks for sharing, and please keep the feedback coming.
@cbrauchli A theme across the reviews seems to be that people want more visibility into which files and symbols the agent actually pulls into context on big codebases, but also a way to steer it. Any plans to expose and let users control context selection more directly ?