API Maker

The backend engine for the AI era

Chapters

What you will learn

  • The System API page
  • Encrypt and decrypt
  • Data checked against a table, and the real shape of a table
  • A system API opened with one setting
  • From a terminal
  • The same from code
In the docsAll system APIs
Transcript606 words · click a time to play from there

Intro

Encryption, hashing, tokens, Redis keys, data checks, indexes and events : APIs every account already has. In this video, the system APIs of API Maker, from the panel, from a terminal and from your code.

Building blocks, the usual way

Common needs usually mean a library to choose for each one, for crypto, tokens and the cache, its keys and its settings wired into every service, and one more endpoint to secure each time. With API Maker, they are ready in every account, with the keys of the secret of the server, and closed until you open one. In this video : the System API page, encrypt and decrypt, data checked against a table, the real shape of a table, a system API opened to applications, and the same from code.

The System API page

APIs and Logic, System API : every system API, each with its address. No-Access : nobody calls it from outside until a setting opens it. Only Get token is public : it is how an application signs in. Test API opens one in API Testing : the panel may call them all.

Encrypt and decrypt

The body : an order to send to a browser, encrypted. Send : the cipher text, made with the transfer key of the default secret. Decrypt data, its body, that text, Send : the order, back.

Checks of a table

API Testing, System APIs, Is valid data for table. The body : a customer of PostgreSQL, with a city given as a name, for a save. Send : nothing is written, and the errors come field by field. Pune is not a number for city_id, and email, loyalty_tier and created_at are required. For an update, only the fields sent are checked : one error left. Get table meta data, its body, the products of MongoDB.

Send : every field, its type, how many of the thirty six products have it, and whether its values repeat. On MongoDB, where nothing declares the shape of a collection, that is the shape it really has.

A system API opened

Now an API for applications : Hash data. Add System API Setting : the template says TOKEN_ACCESS, so an API user whose groups allow it may call it. Save. The chip says AM-User. From a terminal, before the setting : four oh one, you are not authorized to access this API. After it, with the token of an API user : the hash of the e-mail, the same every time.

Without a token : four oh one again.

The same from code

The same blocks from your code : g.sys.system.encrypt and decrypt, hash, isValidDataForTable and getTableMeta, and g.sys.cache for Redis keys with a lifetime. Called from a custom API, they need no setting : your code is trusted. And the same names exist in the eight languages, each in its own case.

Good to know

Execute plain query is not on the page : your code calls it. Create indexes, Drop indexes and Get indexes work on the indexes of a table. Emit event and Emit event WS start events and WebSocket messages : videos twenty one and twenty two. And a setting can also cache the answers of a system API, or accept only encrypted bodies.

Built on standards

One implementation of each block, for every team and every language. Closed by default, opened one by one, with the same tokens and groups as every other API. So code written by an AI assistant calls g.sys.system.hash, and never writes its own crypto.

Recap

To sum up. Ready-made APIs in every account. Encrypt, decrypt and hash. Data checked against a table, and its real shape. Closed until a setting opens one. And the same from code. Next : utility classes and process initializers. API Maker. Build faster, spend smarter, deliver better.