Vibe coding tools

What is Cursor AI?

Cursor is an AI code editor built on VS Code. What it does, how its 2026 acquisition by SpaceX changed plans and models, and how to use it safely on real code.

Cursor is an AI-powered code editor built on VS Code, made by Anysphere, which SpaceX acquired in a deal that closed in August 2026. It offers AI chat, autocomplete and agents that edit your codebase. Its plans now bundle Grok models, OpenAI models were scheduled to leave Cursor on 12 November 2026, and Anthropic and Google models remain available.

Published 9 November 2026 · Last updated 9 November 2026

By Michal Měřínský. Audit methodology.

What Cursor does

Editor: a fork of VS Code, so extensions, themes and keyboard shortcuts carry over.
Tab completion: predicts the next edit, often across several lines.
Chat and agent: ask about the codebase, or let an agent plan and make changes across files and run commands.
Cloud agents: run tasks in the background, on higher plans.
Rules: project files that tell the AI your conventions.

What changed in 2026

SpaceX announced the acquisition of Anysphere, Cursor's maker, in June 2026, and the deal closed on 14 August 2026, according to news reports.
Cursor's plans were restructured into Free, Plus, Premium, Super, Ultra, Teams and Enterprise, bundling access to Grok models and other xAI products.
OpenAI announced it would wind down its contract, with OpenAI model access in Cursor scheduled to end on 12 November 2026. Reports said Anthropic and Google models remain supported.

If your workflow depends on a particular model, check Cursor's current model list. Availability now depends on business agreements that can change.

Using Cursor safely on real projects

Review diffs before accepting. Agents can change many files at once.
Commit before large agent tasks, so you can roll back.
Keep secrets out of files the AI reads, using ignore settings for .env files.
Give agents command permissions deliberately, and require approval for anything touching production or databases.
Write rules for your security conventions, such as where API keys may be used.

Cursor compared with agents

Cursor keeps you in an editor with the code visible. Terminal agents such as Claude Code and Codex take larger tasks with less hand-holding. Many developers use both: an editor for daily work, an agent for bigger changes.

What a scan can and cannot tell you

WebAuditOS checks what you ship from Cursor, not the editor: security headers, console errors, duplicate titles, framework version leaks and the rest of the deployed site's basics.

It cannot review the diffs an agent made or see your Cursor settings.

AI edits that pass your own click-through can still remove a header or break a route. A scan after each release can reveal regressions within its crawl and browser-sampling limits.

Short answers

Is Cursor free? There is a free plan with limited agent and chat requests.

Who owns Cursor? SpaceX, since the acquisition of Anysphere closed in August 2026.

Can I still use Claude in Cursor? Reports in late 2026 said Anthropic models remain available.

Is Cursor the same as VS Code? It is built on VS Code, with AI features added.

In the product

The checks this maps to

Every check named here runs on every scan, on every plan - what a plan changes is how many pages are crawled, how long the history is kept and what you can do with the result. These are the ids the findings carry, so you can match a line in a report to the explanation above.

sec.headers.csp - Content-Security-Policy header
tech.console-errors - No JavaScript errors in the console
seo.title.unique - Titles are unique across pages
sec.disclosure.powered-by - X-Powered-By header removed

Run a free audit See a full sample report See the whole check library