Dilnaka Cloud is a deployment platform built on AWS that runs your apps and databases as isolated Docker services. Start on a shared instance for small or early projects, then move up to a dedicated Custom instance when your workload outgrows shared capacity.
Deploy in minutes. Connect a GitHub repo or push a Docker image and ship your first app.
Get started →Understand when to stay on shared capacity and when a dedicated Custom instance pays off.
Compare →Sign in from a terminal and work with your projects without leaving the shell.
Use the CLI →Ready-to-run starting points, copied into your own GitHub account on deploy.
Browse templates →Go from sign-in to a running deploy in four steps. Every account starts with a 7-day free trial on shared capacity — no card required.
Sign in with GitHub and start your free trial instantly.
A project groups your apps, databases, and services into one workspace.
Link a GitHub repo that ships a Dockerfile, or point straight at a published Docker image.
Push to ship. Watch build logs, get a public URL, and roll back if needed.
Core concept
This is the choice that shapes how you run on Dilnaka Cloud. Both run your apps as isolated Docker containers — the difference is whether the underlying server is shared with other customers or reserved entirely for you.
Your apps run as isolated containers on shared EC2 capacity. Cheapest way to start, and ideal when you have small or early-stage clients and want predictable monthly cost without paying for a whole server.
A full EC2 server reserved for you alone. Run multiple of your own apps and databases on it with no noisy neighbours. The right move once you have serious or high-potential clients who need guaranteed resources and full isolation.
Any language or framework runs here, because the unit we deploy is a container, not a runtime we have to recognise. Ship a Dockerfile in your repo and Dilnaka builds it on the host. Caspian apps are the one exception: if the repo has no Dockerfile, a matching one is generated for you.
Next.js
Dockerfile
Django
Dockerfile
FastAPI
Dockerfile
Flask
Dockerfile
Laravel
Dockerfile
Rails
Dockerfile
Node / Express
Dockerfile
Go
Dockerfile
Rust / Axum
Dockerfile
Actix
Dockerfile
Static sites
Dockerfile
Connect a repository and every push to the tracked branch builds and redeploys. Pick the branch and root directory, limit which paths trigger a build, and optionally hold the deploy until the commit's GitHub checks pass.
Point at a published image and run it directly — no repository, no build step. Useful for off-the-shelf software and for images your own CI already builds.
A template is a ready-to-run starting point. Deploying one copies the source into your own GitHub account and provisions it, so what you get is your project — not a shared demo.
OpenClaw
Open-source personal AI assistant that really does things — inbox, email, calendar, and more. Deploys from the official container image.
TypeScript OpenClaw Dedicated only
Add a database to a project and it is provisioned with its own credentials, injected into your app's environment, and browsable from the dashboard. Five engines are managed for you.
PostgreSQL
Managed
MySQL
Managed
MariaDB
Managed
Redis
Managed
MongoDB
Managed
Data browser
Browse tables, run SQL, and inspect Redis keys from the project view — no client to install. Queries run as your database's own limited role.
Backups & restore
Take a snapshot on demand or put a database on a schedule with a retention window. A restore boots an isolated copy so your live database is never overwritten.
Encrypted at rest
Credentials and connection strings are stored with envelope encryption, connections use TLS, and each tenant's database is isolated from its neighbours on the host.
dilnaka connect is in progress so
local tools can reach a database without it being exposed to the public internet.
Every app gets a managed subdomain, and you can add your own domain on top. Add the DNS record we show you; once it resolves to your app, the domain goes live and a certificate is issued automatically on the first request.
Your container never listens on the public internet directly — it binds to the host's loopback interface, and traffic reaches it only through the managed proxy.
Read build logs, runtime output, and HTTP request logs per service, watch CPU and memory pressure, and track usage against your plan before it turns into an overage.
The Dilnaka CLI connects a terminal to your workspace. It needs Node.js 18.17 or newer and installs with no other dependencies.
$ npm install -g @dilnaka/cli $ dilnaka login # approve the code in your browser $ dilnaka list # your projects and services $ dilnaka link # tie this folder to a project $ dilnaka status # server, session, linked project
dilnaka login prints a short
code and opens your browser. You approve that code while already signed in, and the CLI
receives a personal access token — no password typed, no token pasted.
Tokens are listed under
Account → CLI Tokens.
Revoking one stops that machine on its next command. For CI, create a token there and
expose it as DILNAKA_TOKEN.
These commands are being built and are not available yet:
dilnaka connect — open a local port tunnelled to one of your databasesdilnaka variables / run — read service variables, run a command against themdilnaka logs / redeploydilnaka up — deploy the current directoryStart free on a shared instance, then move to a dedicated Custom instance whenever your workload is ready for it.