Build with AI. Ship with your tools.

By · Contact: nexusai.run/contact

An AI app builder that fits your dev workflow: generate a full-stack app, edit it in your editor, push to GitHub, and deploy from the CLI, CI, or an agent.

The AI app builder that fits your dev workflow.

Generate a full-stack app in chat, then treat it like any other codebase: pull it into your editor, push it to GitHub, and deploy from the CLI, a CI pipeline, the REST API, or Claude Code, Cursor, and Codex over MCP.

The code is yours from the first prompt.

Pick the model, or bring your own API key. Read and change every file. Deploy with the database and workers the app needs, and keep operating it from the terminal.

Internal tool with Postgres

Generate an admin tool in the AI Builder, pull it to edit locally, and deploy it with a managed Postgres database.

Split frontend and API

Deploy the backend/ and frontend/ folders of one repository as two deployments with --root-dir, the API with its database.

Agent-built app over MCP

Let Claude Code, Cursor, or Codex push files to the AI Builder for a preview, then deploy and watch the logs.

Code you own

Edit every file in the built-in editor or in your own. nexus builder pull and push sync with the same project, and any project pushes to your GitHub repository.

Your model, your key

Run the builder agent on Claude, GPT, Gemini, or Grok. Add your own API key and turns on it do not count against your plan’s AI tokens.

Deploy from anywhere

Ship from the dashboard, the nexus CLI, CI with a scoped NEXUSAI_TOKEN, the REST API, or MCP tools, to NEXUS AI or your own Google Cloud, AWS, or Azure account.

Your toolchain

CLI

Deploy a repository with Postgres and Redis and wait for the result.

nexus deploy source \
  --repo https://github.com/you/app \
  --services postgres,redis \
  --wait

CI

Redeploy from GitHub Actions with a scoped, expiring token.

- run: npm install -g nexusapp-cli@latest
- run: nexus deploy redeploy app --yes --wait
  env:
    NEXUSAI_TOKEN: ${{ secrets.NEXUSAI_TOKEN }}

MCP

Give Claude Code 74 OAuth-scoped tools for deploys, logs, and databases.

claude mcp add --transport http nexus-ai https://mcp.nexusai.run/mcp

AI Builder in your editor

Pull an AI-built project, edit it locally, and push the changes back.

nexus builder pull -p <project-id> --out ./app
nexus builder push -p <project-id> --from ./app
nexus builder deploy -p <project-id> --name app --provider docker

Frequently asked questions

Can I use my existing development workflow?

Yes. Import a GitHub repository or push the project to GitHub, edit locally with nexus builder pull and push, and deploy from the CLI, CI, the REST API, or MCP tools.

Can I bring an existing repository?

Yes. Import a GitHub repository into the AI Builder for a running preview, or deploy it directly with nexus deploy source. For a repo with separate frontend and backend folders, deploy each one with --root-dir.

Which frameworks are supported?

The AI Builder generates Next.js apps and adds Prisma when the app stores data. Deployments detect Node.js, Next.js, React, Vue, Angular, Vite, Python, Django, FastAPI, PHP, Laravel, Symfony, Ruby on Rails, and static sites, or use your own Dockerfile.

Can I use my own model or API key?

Yes. The builder agent runs on Claude, GPT, Gemini, or Grok. Add your own Claude, OpenAI, or Gemini key on any plan, or a Grok or OpenRouter key on a paid plan; turns on your key do not use the plan’s AI tokens.

What can a deployment include?

On the NEXUS AI runtime: the app plus PostgreSQL, MySQL, MongoDB, or Redis, background workers, persistent volumes, and S3-compatible buckets. On Google Cloud Run, AWS App Runner, and Azure Container Apps: a single container with a managed PostgreSQL or MySQL database. Every deployment gets logs, health checks, custom domains, and backups.

How are changes verified before they ship?

The builder agent runs project checks and fixes the errors it finds. Full-stack Next.js apps also run in a development sandbox with their dependencies and a database, and build errors in the preview trigger automatic fix attempts.

Is there a CLI and an API?

Yes. The nexus CLI and the REST API cover deployments, logs, databases, backups, storage, secrets, and domains, and the MCP server exposes the same operations to AI agents as 74 tools.

How do CI pipelines authenticate?

With an nxk_ access token in the NEXUSAI_TOKEN environment variable. Create it with nexus token create, choose only the scopes the pipeline needs, and set an expiry.

Can I take my code elsewhere?

Yes. Every project is standard code you can push to your own GitHub repository, and deployments are standard containers.

Can it run in my own cloud?

Yes. Connect your own Google Cloud project on Starter, or AWS, Google Cloud, or Azure on Pro, and apps and databases deploy into that account. NEXUS AI Enterprise runs the whole platform in your cloud account or on-premises.

About NEXUS AI

NEXUS AI is an agentic AI app builder and full-stack deployment platform. Explore the AI App Builder, learn more on the About page, read the documentation, or contact the team through nexusai.run/contact.

Start for free · Read the documentation · About NEXUS AI · Contact