Everything Your Cloudlet Can Do, In One Place

Sign in to any backend

The dashboard talks to any Magic backend. The selector remembers every cloudlet you have signed into and keeps a separate JWT for each, so switching between them never means signing in twice.

A magnetic link signs you in from an email when SMTP is configured, and Continue with Google authenticates over OIDC against whatever providers your backend has registered.

Your cloudlet at a glance — and as an AI agent

Version, endpoints, users, tasks and log items in one row. Below them, the MCP banner hands any MCP-capable agent your endpoints as callable tools — point Claude, or anything speaking the protocol, straight at it.

The Chatbot Wizard runs the other way: give it a website, and the backend crawls it, turns what it finds into training data, and hands you an embeddable chatbot grounded in your own content.

Edit and run anything on your server

A file tree over the whole backend, with files open as tabs — several at once, each tracking whether it is dirty. The editor colours Hyperlambda correctly and completes against the backend's actual vocabulary with Ctrl+Space.

Endpoint files are invoked as real endpoints with arguments and response headers; everything else is evaluated directly. Upload and download files, preview HTML, images, JavaScript and CSS, and press F1 to ask the AI what a piece of code does.

Run Hyperlambda without saving anything

Input on the left, result on the right, F5 to run. It is the scratchpad you reach for when you want to know exactly what a slot returns — and it saves snippets when a scratch turns out to be worth keeping.

Query your databases — and design them

The SQL tab is a query editor with syntax colouring, completion over your real tables and columns, saved snippets, .sql import, CSV export, and a Safe mode that caps what a careless query can do. Selecting text executes only the selection.

The Designer tab renders every table as a card — keys, types, nullability and foreign keys spelled out — and lets you add tables, columns and relations without writing a line of DDL.

SQLite in, and everything else connected

Create SQLite databases, back them up, restore from an uploaded backup, and drop the ones you no longer need. On the second tab, connect to external MySQL, PostgreSQL and SQL Server instances — after which the whole dashboard treats them exactly like a local database.

Tables in, secured API out

Point it at a database, tick the tables, and it writes a complete CRUD backend: POST, GET, PUT and DELETE per table, plus paging, sorting, aggregates, distinct and search endpoints on request.

You choose which roles may call what, whether writes are logged, how long GET responses may be cached, and whether existing files may be overwritten. The second tab wraps a custom SQL statement in an endpoint of its own.

Every endpoint, invokable with real arguments

Grouped by module, with its HTTP verb. Open one and you get a form built from its actual arguments — fill them in, invoke it, and read the status code, response headers and body rendered by content type.

Endpoints that accept multipart get a file picker with inline image thumbnails; endpoints that return files offer them as downloads. Each module will also hand you its OpenAPI specification.

The authorisation surface, in one table

Who may reach your backend, and what they may do. Create users, assign roles, change passwords, lock accounts. Roles are the same objects the Generator gates endpoints with and the same ones inside the JWT — so this one screen is the entire access model.

Hyperlambda on a schedule

Backups, cleanup jobs, crawls and alerts, run on demand or on a schedule. Tasks can repeat on an interval, fire once at a fixed date, or follow a custom pattern — and you can execute any of them immediately, which asks first, because a task named delete-old-log-items means it.

Models, training data and chatbots

The largest screen in the application. Models define which LLM answers, at what temperature, with what system message and threshold. Training data is what they answer from — crawled, uploaded or hand-written.

Vectorising turns it into embeddings; embedding hands you the snippet that drops the finished chatbot onto a web page. A history tab shows what people actually asked it.

The Bazar, one click away

MCP, OAuth, OpenAI, Ollama, HuggingFace, Shopify, HubSpot, NetSuite, scraping, charts and a couple of dozen more. One click installs a module into your cloudlet, endpoints and all — or install your own from a ZIP file.

Your account, and long-lived tokens

Your name, your email, your password — and access tokens, which is the part that matters operationally. Generate a long-lived JWT for a service account, a CI pipeline or an integration, scoped to the roles you pick and expiring on a date you choose.

What your backend has been doing

Newest first, filterable, with severity on every row. It is where you go when something did not work — and the generated CRUD endpoints write to it automatically when you ask them to log writes.