Cursor: loved for UX, held back by reliability.
766 Hacker News opinions from 600 developers, Nov 2025 – Nov 2026. 583 come from people who say they use it. Here is what they praise, what they complain about, and who switched.
Where Cursor is strong — and where it is not.
Cursor earns 55% approval across 766 opinions. Developers praise its UX, while reliability is the recurring complaint. In the last 90 days it sits at 53%.
Pick it if you
- Want Claude, Codex and other agents in one interface
- Care most about autocomplete speed
- Want to switch models inside the editor
Look elsewhere if you
- Need dependable agent behavior
- Have a tight budget or hit usage caps
- Depend on extensions Cursor does not support
Where it wins and loses.
Mentions across 766 opinions. Bars are counts, not votes; a tool can be both praised and criticized on the same topic.
View data table
| Topic | Praise | Complaints | Mentions |
|---|---|---|---|
| UX | 101 | 59 | 160 |
| Speed | 44 | 8 | 52 |
| Customization | 21 | 9 | 30 |
| Code quality | 59 | 49 | 108 |
| Autonomy | 30 | 22 | 52 |
| Cost & pricing | 58 | 60 | 118 |
| Usage limits | 17 | 20 | 37 |
| Reliability | 14 | 58 | 72 |
How opinion has moved.
Monthly approval, Nov 2025 – Oct 2026. Bars show opinion volume; low-volume months are still included but faded.
View data table
| Month | Approval | Opinions |
|---|---|---|
| Nov 2025 | 57% | 50 |
| Dec 2025 | 56% | 68 |
| Jan 2026 | 56% | 100 |
| Feb 2026 | 61% | 76 |
| Mar 2026 | 54% | 77 |
| Apr 2026 | 49% | 118 |
| May 2026 | 54% | 51 |
| Jun 2026 | 58% | 92 |
| Jul 2026 | 64% | 25 |
| Aug 2026 | 51% | 73 |
| Sep 2026 | 70% | 20 |
| Oct 2026 | 44% | 16 |
What developers say.
Summaries are paraphrased. Each links to the original Hacker News comment.
Praise
4 selectedCursor is convenient because it puts Claude, Codex and other agents in one interface.
Moved all repos to Cursor Origin after an outage, calls it fast, and recommends others try it.
Uses Cursor Pro heavily, obtained through a bundle subscription.
Likes Cursor's review UI and that it lets them pick any model.
Complaints
4 selectedCursor's missing support for wanted extensions frustrated them, so they stopped using it.
Cursor got worse as it added agentic features, after they loved its tab-completion era.
Considering cancelling the Cursor subscription because Claude models work better for them.
Cursor agents run for hours and interrupt their work, so they want cloud execution.
Questions developers ask.
Is Cursor worth it?
Is Cursor better than Claude Code?
Why do developers stop using Cursor?
The data behind this page.
Developers give Cursor a 55% approval score across 766 opinions from 600 authors, and 53% over the last 90 days. They like having Claude, Codex and other models in one interface, plus fast autocomplete and a useful review UI. Reliability is the weak point, with 14 positive mentions against 58 negative, while cost is split at 58 positive and 60 negative. The recent trend is roughly flat, with monthly approval swinging from 70% in September to 44% in October on small samples.
Every number comes from public Hacker News comments collected through the HN Search API. A language model tagged each comment for sentiment, first-hand use, topic and switches; a human spot-checks samples. No vendor can pay for placement.
Get the monthly report
Which AI coding tools gained or lost developer approval on Hacker News, and why people switched. One email a month.