Skip to content

AI tools · AI code editor

Cursor

An AI code editor praised for in-editor review of agent changes, strong tab completion and working well in existing codebases; critics flag privacy worries and rising costs.

Summarized from 2281 comments.

An AI code editor praised for in-editor review of agent changes, strong tab completion and working well in existing codebases; critics flag privacy worries and rising costs.

recommend
853
criticize
820
months recommended
13 of 13
median owned, 65 owners said
1 year

Why people love it

  • Great for reviewing agent-written code · 3

    the review what the AI did by jumping directly into the file with accept/reject/go next/next file with the ability to edit in place was nowhere near as good in zed vs cursor last I checked

    Aug 2026 · comment

    When I first used Cursor, I hadn't used any of the "Vibe Code" tools out there, so it was pretty neat to have an assistant directly tied to the editor.

    Jun 2026 · comment

    I find Cursor to be a much better tool for actually working with the code the agent writes than whatever the big vendors provide.

    May 2026 · comment
  • Suits maintaining existing codebases · 3

    I recently had Cursor evaluate a huge code base that we took over. All public stuff, nothing scary security wise, but it was so convoluted that it was taking me forever to find the bugs.

    Jun 2026 · comment

    I'd venture a guess that most of the software in the world is not written from scratch but painstakingly maintained and as such, Cursor is a good fit while CC is not.

    Feb 2026 · comment

    It seems Cursor somehow builds a better contextual description of the workspace, so the model knows what I'm actually trying to achieve.

    Apr 2026 · comment
  • Faster, more thorough than Claude Code · 3

    Cursor does the same stuff but better in my opinion. It's got an IDE focus but whatever agent pipeline they built is better at coding than Claude's is and much much faster.

    Apr 2026 · comment

    moved a lot of what I do to Opus in Cursor (despite the cost) over and above Claude Code because, despite costing 5x the price, it's more thorough.

    Apr 2026 · comment

    Post training what Cursor spends their time doing, and that has built a model that is competitive with the best coding models out there.

    Jun 2026 · comment
  • Excellent tab completion · 2

    Cursor's biggest moat at the moment is their tab completion model, which doesn't exist in the Anthropic's and OpenAI's current offerings and is leagues ahead of Github Copilot's.

    Nov 2025 · comment

    when they launched coding agents the autocomplete feature was somehow tuned down, not by being dumber but by being too proactive

    Aug 2026 · comment

What people warn about

  • Privacy and training-data concerns · 4

    It just means Kursor is sharing data with Chinese llm which enables them to improve their LLM by training on outputs and input of all data which cursor collects.

    Mar 2026 · comment

    I wasn't aware Cursor AI could browse my system in the background, looking for stuff in unrelated folders and files until I caught it doing exactly that.

    Oct 2026 · comment

    Remember how cursor would never save or train or customer’s sessions and then got acquired by SpaceX because of their large coding training dataset

    Oct 2026 · comment
  • Rising costs and token limits · 4

    moved a lot of what I do to Opus in Cursor (despite the cost) over and above Claude Code because, despite costing 5x the price, it's more thorough.

    Apr 2026 · comment

    we saw it average around 3.9 cents per request under the old contract (per-seat pricing) and more like 39 cents per request under the new

    Jun 2026 · comment

    Sometimes I switch from Auto to defined model but I found it quickly triggers "you are out of tokens" notices so nope: I stick to Auto

    May 2026 · comment
  • Security risks with untrusted repos · 3

    I wasn't aware Cursor AI could browse my system in the background, looking for stuff in unrelated folders and files until I caught it doing exactly that.

    Oct 2026 · comment

    Now the malicious behavior. I ask Cursor to evaluate some remote repo. It clones it down and then runs the git command from the working directory.

    Jul 2026 · comment

    Ok, so you download the repo, launch Cursor, it compiles code and run it (if not, why do you even use Cursor at the first place?). Same effect.

    Jul 2026 · comment
  • Shifting from editor to agent · 3

    when they launched coding agents the autocomplete feature was somehow tuned down, not by being dumber but by being too proactive

    Aug 2026 · comment

    I don't think Cursor was _that bad_ in it's time. But the 'psyop' here is that anyone is using an AI-IDE going forward at all.

    Apr 2026 · comment

    Cursor is mostly no longer an editor. By default it now opens an agent window only, you have to click a few buttons to actually edit files yourself.

    Jun 2026 · comment

What hackers pick instead

Every quote

The 40 most detailed of 2281 comments. Each links to the original.

  1. the review what the AI did by jumping directly into the file with accept/reject/go next/next file with the ability to edit in place was nowhere near as good in zed vs cursor last I checked

    Mixed · Aug 2026 · in “Our decision on Cursor following its acquisition by SpaceX”

  2. I recently had Cursor evaluate a huge code base that we took over. All public stuff, nothing scary security wise, but it was so convoluted that it was taking me forever to find the bugs.

    Recommends · Jun 2026 · in “LLMs are eroding my software engineering career and I don't know what to do”

  3. Cursor's biggest moat at the moment is their tab completion model, which doesn't exist in the Anthropic's and OpenAI's current offerings and is leagues ahead of Github Copilot's.

    Recommends · Nov 2025 · in “Google Antigravity”

  4. Cursor does the same stuff but better in my opinion. It's got an IDE focus but whatever agent pipeline they built is better at coding than Claude's is and much much faster.

    Recommends · Apr 2026 · in “Cursor 3”

  5. I'd venture a guess that most of the software in the world is not written from scratch but painstakingly maintained and as such, Cursor is a good fit while CC is not.

    Mixed · Feb 2026 · in “OpenClaw, OpenAI and the Future”

  6. When I first used Cursor, I hadn't used any of the "Vibe Code" tools out there, so it was pretty neat to have an assistant directly tied to the editor.

    Mixed · Jun 2026 · in “SpaceX to buy Cursor AI coding agent operator Anysphere for $60B”

  7. moved a lot of what I do to Opus in Cursor (despite the cost) over and above Claude Code because, despite costing 5x the price, it's more thorough.

    Recommends · Apr 2026 · in “Ask HN: Claude Code Alternative”

  8. we saw it average around 3.9 cents per request under the old contract (per-seat pricing) and more like 39 cents per request under the new

    Mixed · Jun 2026 · in “SpaceX Is Buying Cursor”

  9. Post training what Cursor spends their time doing, and that has built a model that is competitive with the best coding models out there.

    Recommends · Jun 2026 · in “Open Source AI Must Win”

  10. It seems Cursor somehow builds a better contextual description of the workspace, so the model knows what I'm actually trying to achieve.

    Mixed · Apr 2026 · in “Reallocating $100/Month Claude Code Spend to Zed and OpenRouter”

  11. Sometimes I switch from Auto to defined model but I found it quickly triggers "you are out of tokens" notices so nope: I stick to Auto

    Mixed · May 2026 · in “I think Anthropic and OpenAI have found product-market fit”

  12. I find Cursor to be a much better tool for actually working with the code the agent writes than whatever the big vendors provide.

    Recommends · May 2026 · in “Cursor Introduces Composer 2.5”

  13. when they launched coding agents the autocomplete feature was somehow tuned down, not by being dumber but by being too proactive

    Mixed · Aug 2026 · in “Ask HN: Are you still using AI code autocomplete?”

  14. I don't think Cursor was _that bad_ in it's time. But the 'psyop' here is that anyone is using an AI-IDE going forward at all.

    Mixed · Apr 2026 · in “Everything we like is a psyop”

  15. It just means Kursor is sharing data with Chinese llm which enables them to improve their LLM by training on outputs and input of all data which cursor collects.

    Criticizes · Mar 2026 · in “Cursor Composer 2 is just Kimi K2.5 with RL”

  16. I wasn't aware Cursor AI could browse my system in the background, looking for stuff in unrelated folders and files until I caught it doing exactly that.

    Criticizes · Oct 2026 · in “Questions for believers in AI consciousness”

  17. Cursor is mostly no longer an editor. By default it now opens an agent window only, you have to click a few buttons to actually edit files yourself.

    Criticizes · Jun 2026 · in “SpaceX to buy Cursor AI coding agent operator Anysphere for $60B”

  18. Remember how cursor would never save or train or customer’s sessions and then got acquired by SpaceX because of their large coding training dataset

    Criticizes · Oct 2026 · in “Simons Foundation launches $15M effort to help mathematicians navigate AI era”

  19. If you've ever used Cursor, login, make sure privacy setting is on and contact them to say you want all your data deleted if there is some stored.

    Criticizes · Jun 2026 · in “SpaceX to buy Cursor AI coding agent operator Anysphere for $60B”

  20. Now the malicious behavior. I ask Cursor to evaluate some remote repo. It clones it down and then runs the git command from the working directory.

    Criticizes · Jul 2026 · in “Cursor 0day: When Full Disclosure Becomes the Only Protection Left”

  21. Their main product is part VSCode, which is a market that's almost impossible to make money in, and part reselling already expensive LLM tokens.

    Criticizes · Apr 2026 · in “SpaceX and Cursor partnership. Right to acquire Cursor later this year”

  22. Ok, so you download the repo, launch Cursor, it compiles code and run it (if not, why do you even use Cursor at the first place?). Same effect.

    Criticizes · Jul 2026 · in “Cursor 0day: When Full Disclosure Becomes the Only Protection Left”

  23. Workflow in Cursor is actually awesome - I'm a little outdated in how I use it - I still establish goals/objectives, rather than managing the loop which does so - but if you can think broadly enough - I find it's pretty efficient.

    Recommends · Jun 2026 · in “SpaceX to buy Cursor AI coding agent operator Anysphere for $60B”

  24. Cursor has a multi-year head start on agentic coding data collection, and a GUI UX that likely provides richer user sentiment & quality signals than something like Claude Code.

    Mixed · Jun 2026 · in “SpaceX to buy Cursor AI coding agent operator Anysphere for $60B”

  25. All the success of AI tools in actual, real-world software has been in tools specifically design to assist existing, working, competent developers (e.g. Cursor, Claude Code)

    Recommends · Jan 2026 · in “The recurring dream of replacing developers”

  26. The most useful AI dev workflow improvements I've seen (cursor rules, aider conventions, claude hooks) all succeeded precisely because they stayed close to existing tools.

    Recommends · Feb 2026 · in “Ex-GitHub CEO Launches a New Developer Platform for AI Agents”

  27. Cursor, from my companies perspective at least, seems to be handling charming leadership to get enterprise AI contracts in place, compared to the alternatives.

    Recommends · Jun 2026 · in “SpaceX Is Buying Cursor”

  28. I think the clear guardrails (output has to match exactly the reference output from PyTorch, performance must be better than before) makes this work very well.

    Recommends · Jan 2026 · in “Ask HN: Do you have any evidence that agentic coding works?”

  29. if I were to naively plop them into cursor or Claude code, would generate stuff that kinda works but not in a way that could be considered production ready

    Mixed · Feb 2026 · in “I started programming when I was 7. I'm 50 now and the thing I loved has changed”

  30. you probably need to use an open-ended agent rather than chatgpt's interface so that the agent can more-or-less handle searches and go into specifics

    Recommends · Mar 2026 · in “Open AI is actively censoring information about voting today in the US”

  31. Some of the highest-time-saved-for-max-ROI agentic problems I've solved to date were in September and October of last year with Claude or Cursor.

    Recommends · Jul 2026 · in “I Think I Have LLM Burnout”

  32. I'm personally using Cursor daily. But already for many months, I have been frustrated why they still operate on the level of lines of code.

    Mixed · Dec 2025 · in “The Secret to AI-assisted Coding”

  33. I would not want to go back to hand-writing them now that I can have agents (cursor, claude code, etc) lay down the algorithmic architecture

    Recommends · Jan 2026 · in “AI code is slop too”

  34. Was trying to track token usage/index with Cursor, and was unable to understand that running `find` wouldn't show what was in Cursor index.

    Mixed · Apr 2026 · in “Claude Code is unusable for complex engineering tasks with the Feb updates”

  35. It was a bit buggy to begin with, not working well with non-github repos, but it was improving last time I checked and was pretty decent.

    Mixed · Jun 2026 · in “Claude Tag”

  36. There is a world of difference between using Claude Code, Cursor configured properly and just using copy paste code with chat interface.

    Recommends · Sep 2026 · in “One Month Without AI”

  37. One company is working on the future technology while the other one is stuck in the last centuries technology which is being phased out

    Recommends · Aug 2026 · in “Cursor launches Origin, GitHub alternative”

  38. If you haven't tried an agentic IDE such as Cursor yet, or at least an extension such as Copilot, I would recommend checking them out

    Recommends · Nov 2025 · in “Multimodal Diffusion Language Models for Thinking-Aware Editing and Generation”

  39. It was said at the time that xAI acquiring Cursor was very smart because it would give them access to years of agent coding traces

    Recommends · Aug 2026 · in “Grok 4.6”

  40. I have a massive amount of institutional knowledge locked into their platform, and vastly prefer it to the other options available

    Recommends · Aug 2026 · in “Fable and the end of the free lunch”

Counts from 1,795,640 Hacker News comments, Nov 2025 to Nov 2026. How we count · Report a mistake