The runtime your agent deploys to

Your AI built the app. Skip the DevOps.

No repos, Dockerfiles, env vars, or DNS. One call and it goes live — a real container with a Postgres database, file storage, and a public URL, wired in and deployed by your agent over MCP. No walled garden, no lock-in.

Start Free Read the docs

No credit card required
Flat pricing — everything included
Live in seconds

Works in Claude Code, Cursor, Codex, Cline, and any MCP client.

The difference

Two ways to ship what your AI builds

The usual way

~~Create a repo~~
~~Pick a host, provision a server~~
~~Write a Dockerfile~~
~~Provision a database and storage~~
~~Configure env vars and secrets~~
~~Set up DNS and HTTPS, deploy, then test~~

With BasicDeploy

Your agent makes one call
It's live at a real public URL
Container, Postgres, storage and env — already wired in

AI connector

Deploy without leaving the chat window

Connect BasicDeploy to your AI assistant. Describe your app and it creates the container, deploys your code, and hands back a live HTTPS URL — right inside the conversation. No dashboard, no CLI.

Claude →ChatGPT →

https://mcp.basicdeploy.com/mcp

Add this URL as a connector in Claude or ChatGPT — sign in once, done.

Create your app and deploy right away.

Simple as hell

You don't need to know how any of it works

No servers to manage. No config files. No YAML. No Dockerfile required. Create a container and your app is live at a public HTTPS URL — with a Postgres database and file storage already wired in. That's the whole workflow.

No ~~servers~~
No ~~config files~~
No ~~YAML~~
No ~~DevOps~~
No ~~infra knowledge~~

Create it → it's live.

Quick start

Start in one paste

Give your agent the config below, or call the API directly. A live URL, Postgres and S3 come provisioned — no setup.

# Install the MCP server (or let npx fetch it on demand):
npm i -g basicdeploy-mcp@latest

# Then add this to your agent's MCP config:
{
  "mcpServers": {
    "basicdeploy": {
      "command": "npx",
      "args": ["-y", "basicdeploy-mcp@latest"],
      "env": {
        "BASICDEPLOY_API_KEY": "bd_your_api_key",
        "BASICDEPLOY_URL": "https://basicdeploy.com"
      }
    }
  }
}

Get an API key Read the docs

Features

The persistent runtime your agent deploys to

A database, file storage and a public URL, built into every app and included in the price. No servers, no config — create it and it's live.

Neutral by design

MCP-native and open. Point Claude Code, Cursor, or your own agent at it — one config, no ecosystem lock-in.

Batteries included

Every app gets Postgres, S3-compatible storage, a public URL, and SSH — provisioned and auto-wired into the container. The app just runs.

Connect directly to Postgres & S3

Your database and object storage aren't locked inside the container. Connect straight to your Postgres from pgAdmin, DBeaver or any client, and to your S3 bucket from any S3 tool — same credentials, over a public endpoint.

Share with anyone

Share a running container by email. The recipient gets a link that opens just that container as a guest — no account, no setup. Access is time-limited or permanent, from the dashboard or an agent over MCP.

Build and ship in place

The container is a persistent, live dev environment. Your agent writes code, runs it, and iterates inside it over MCP — exec, terminal, or SSH — live the whole time at its public URL. Deploying is one call over MCP or REST — no YAML, no pipelines. A runtime you build in, not a walled app builder.

Auth for your users

Add sign-up and login to any app you build — hosted pages over standard OpenID Connect, on a fully isolated tenant. Your users, your app, no auth code to maintain.

Deploy from the chat window

Add BasicDeploy as a connector in Claude or ChatGPT — one URL, sign in once, no install or API key. Create your app and deploy it right away, without leaving the chat.

How it works

From agent to live in three steps

No tutorials, no config files. Connect an agent and ship.

01
Connect your agent
Add the BasicDeploy MCP server, or an API key. Any agent, any client.
02
Build and ship
The agent creates a container and works right inside it — writing code, running it, iterating over exec, terminal, or SSH. Database and storage are wired in, and it’s live the whole time.
03
Share the URL
Your app is live at a public URL in seconds — send it to anyone. Or share the running container itself: invite someone by email and they open it as a guest through a link — no account, time-limited or permanent.

How we compare

One flat price. Everything included.

Most platforms start cheap, then meter you — per gigabyte, per request, per seat. BasicDeploy is one flat price with the database, storage, bandwidth and URL already in it. No usage bills, no surprises.

	BasicDeploy that's us	Vercel	Supabase	Railway
Pricing model	Flat, all-in	Per-seat + usage	Base + usage	Per-workspace + usage
Free to start	$0 · no card	$0 · personal use	$0 · pauses when idle	$0 · limited credit
Paid plans from	$12/mo	$20/user	$25/mo	$20/mo
Predictable bill — no usage overages	Yes	No	Capped	No
Database, storage & bandwidth in the price	Yes	No	No	No
PostgreSQL database included	Yes	Partner, metered	Yes	Metered add-on
Object storage included	Yes	Metered	Metered	Metered
Persistent container, always-on optional	Yes	No	No	Yes
Public HTTPS URL + custom domain	Yes	Yes	Paid add-on	Yes
Deploy from your AI agent over MCP	Native, primary	Add-on	Management only	Wraps CLI
Hosted auth (OIDC) built in	Yes	No	Yes	No

Competitor plans and prices from each vendor's own pricing pages, August 2026. All four have a free tier, and all offer real object storage and public URLs — BasicDeploy's difference is bundling the database, storage and bandwidth into one flat, predictable price. Railway runs persistent containers too; Supabase includes Postgres and hosted auth.

Pricing

Flat pricing. Everything included.

One flat price per container — no usage or per-resource charges, ever. The database, object storage, bandwidth and public URL are included on every plan. Start free.

Free

$0forever

Perfect for trying things out

3 containers
Up to 256 MB each
1 GB storage
Sleeps when idle
PostgreSQL + S3 included
Auth for your users included
Public URL
Community support

Get Started Free

Popular

Pro

$12/mo

For builders who ship often

3 containers
Up to 1 GB each
5 GB storage
Sleeps when idle · free always-on
PostgreSQL + S3 included
Auth for your users included
1 custom domain (bring your own Cloudflare)
Priority support

Upgrade to Pro

Scale

$39/mo

For teams running in production

10 containers
Up to 2 GB each
10 GB storage
Sleeps when idle · free always-on
PostgreSQL + S3 included
Auth for your users included
10 custom domains (bring your own Cloudflare)
Priority support

Upgrade to Scale

Enterprise

Custom

Everything in Scale, plus:

Dedicated, isolated database & storage
Adjustable containers, memory & storage
SSO/SAML + team roles
99.9% uptime SLA + priority support
Custom domains, audit logs, DPA/MSA

Talk to us

Need more? Additional always-on containers $4/mo each.

Give your agent a place to ship

Database, object storage and a public URL included — flat pricing, no usage bills, no DevOps. Start free — no credit card.

Deploy Your First App

Tutorials

Watch it build real apps

Each of these was created and deployed on BasicDeploy from a single prompt. Pick one and see how it is done.

Guestbook app — Next.js + Postgres + S3

One prompt builds a full-stack Next.js guestbook with a PostgreSQL database and S3 image uploads, deployed live over MCP.

Next.js
PostgreSQL
S3
MCP

Fitness tracker — Python REST + web

A quick Python fitness tracker: a REST backend, a web page to log activities, and photo uploads to S3 — created and deployed from a single prompt.

Python
REST API
S3
MCP

ERP/CRM app — 3 containers (frontend + backend + scheduler)

One prompt scaffolds a small ERP/CRM across three containers — a frontend, a REST backend, and a scheduler — with demo products seeded and full CRUD on every record, deployed live over MCP.

3 containers
REST
CRUD
MCP

View all tutorials →