Skip to content

Product

An AI-native development platform, end to end.

Prompt-to-app generation, a real editor, isolated runtimes, databases, GitHub, and deployment in one workspace. Built so the generated code stays yours.

The workspace

Editor, agent, preview, and logs side by side.

Interface previewA static illustration of the Codexa workspace layout, built from the product's UI components. Not a screenshot of a live session.

How the agent works

  1. Step 1: Plan

    The agent reads your prompt and the project files, then writes a short plan of the files it will create or change.

  2. Step 2: Generate

    It edits real files with tools (read, write, rename, delete). Every change is recorded, and a snapshot is taken first.

  3. Step 3: Validate

    Type-check, build, and tests run in an isolated sandbox. Errors are read back and repaired, bounded to two attempts.

  4. Step 4: Run

    Start a preview in the sandbox with streamed logs, viewport controls, and a share link.

  5. Step 5: Ship

    Deploy to Codexa static hosting or Vercel, keep deployment history, and roll back in one click.

Requires an AI provider

AI features require an Anthropic or OpenAI API key configured by the operator of your Codexa instance. When no key is configured the agent shows an explicit "not configured" state instead of pretending to work.

Frequently asked questions

Do I own the code?

Yes. Projects are ordinary files stored in your workspace. You can download them, push them to GitHub, or deploy elsewhere at any time.

Which stacks are supported today?

Starter templates cover React + Vite + TypeScript + Tailwind single-page apps, static HTML/CSS/JS sites, and Node + Express REST APIs. You can import other repositories from GitHub, but the sandbox is tuned for these three.

Does the AI work out of the box?

Only when the operator of the Codexa instance has configured an Anthropic or OpenAI API key. Without one, AI features show a "not configured" state. Everything else (editor, runtime, deployment) works without AI.

What happens when the AI makes a mistake?

A snapshot is taken before every run, each run records its plan, tool calls, and changed files, and you can revert the run with one click. Validation (type-check, build, tests) runs in the sandbox and the agent repairs errors at most twice before handing control back.

Where does my code run?

Never inside the web application. A separate runner service executes projects as an unprivileged user inside Linux namespaces with cgroup memory and process limits and timeouts. Docker and E2B drivers are available for operators who prefer them.

Can my team edit the same file at the same time?

Not yet. Realtime presence shows who is viewing which file, and snapshots make it easy to merge work, but concurrent editing (CRDT) is on the roadmap.

How do you charge?

Free, Pro, Team, and Enterprise plans in INR or USD, monthly or annual. Limits (AI requests and tokens, runtime minutes, deployments, storage, projects, members) are enforced on the server. See pricing.

Start building with Codexa

Free plan, no card required. Describe your idea and get a real project you can edit, run, and deploy.