Cursor
Cursor is an AI code editor that pairs a VS Code-style workspace with chat, agents and a command-line tool for writing and reviewing software. It is mainly used to build features, fix bugs and ship projects by describing tasks in plain language.
#7 in Coding & Development ToolsAI Coding AssistantsFree plan
Who it's for
Best for
Engineering teams that want autonomous agents for codebase tasks and reviews
Developers who want a VS Code workspace with agents and model choice
Not for
Freelancers on a fixed budget who need predictable monthly costs
Teams that need human billing support and refunds on annual plans
Free plan$0
Plus$8/mo
Premium$30/mo
Super$100/mo
Ultra$200/mo
Teams$30/seat/mo
Features
Agentic development
Tasks are handed off to Cursor agents that research the codebase, plan the change and write the code, leaving decisions with the developer.
Cloud agents
Cloud agents run on their own computers to build, test and demo features end to end, working autonomously and in parallel.
CLI agent
A terminal agent brings the assistant outside the editor, with slash commands for actions and @ references for pointing at files.
Codebase indexing
Secure codebase indexing and semantic search give the editor context about project files, so questions and edits stay grounded in the repository.
Context-aware completions
Context-aware completions predict the next edit from the surrounding code, working alongside the chat and agent panels in the editor.
What users say
Pros
- A single prompt in an empty folder produced a working, published Obsidian plugin with releases and icons.
- Switching models mid-task, such as implementing with Opus then reviewing with another provider's model, works in one session.
- Auto-review run mode keeps agent sessions moving without the repeated permission prompts that interrupt other tools.
- Cloud agents handle automated investigations of alarms and requests unattended, becoming part of a team's daily workflow.
- Bugbot code review catches subtle bugs in pull requests before they reach the main branch.
Cons
- Included quota runs out in about five days of normal professional work, leaving the rest of a prepaid month unusable.
- On-demand charges appear without any spending alert, even after the setting was explicitly disabled.
- Billing tickets and account blocks get automated AI replies with no human review and no refund.
- Recent updates drop the editor to about 5 FPS, and the GPU lag makes it unusable for real work.
- Agents change code outside the requested scope and break features that were explicitly marked as preserved.
- Updates remove working features and rearrange the interface, taking away index scoping and automatic worktree management.
A selection of the review sources we used:
Summarized from public user reviews and community discussions. View details
Global Traffic Rank
Top 10k
Alternatives
Rank#7 in category#1 in category#2 in category#3 in category
Free planYesYesYesYes
ProShips working projects from prompts, model switching, cloud agentsShips live apps, large analyses, fast tools, custom hooksHard bugs solved, full projects built, cheap entry plan, editor fitInline autocomplete, multi-repo agents, repository instructions