@vektorapp/api

A zero-dependency, read-only TypeScript client for consuming Vektor content from a website or server-rendered application.

import { createVektorClient } from "@vektorapp/api";

const vektor = createVektorClient({
  accessToken: process.env.VEKTOR_ACCESS_TOKEN,
});

// Uses http://localhost:8080 when baseUrl is omitted.
const page = await vektor.getDocumentBySlug("my-space-id", "about");

Available operations are listSpaces, listDocuments, listDocumentsByCategories, getDocument, getDocumentBySlug, getRevision, listCategories, and search. The client only issues GET requests. For a remote instance, pass baseUrl explicitly.

getDocument and getDocumentBySlug both accept an id or a slug, for the space and the document alike. By default they return the published revision; pass { draft: true } to read the current draft instead, which requires an editor-scoped token.

Properties

A document property holds either one value or a list of them, so document.properties[key] is typed string | string[]. Use the exported helpers rather than assuming a string:

import { propertyScalar, propertyText } from "@vektorapp/api";

propertyText(document.properties.tags); // "docs, guide"
propertyScalar(document.properties.headerImage); // "https://…/hero.jpg"

Vektor returns document content as HTML. Only render it as raw HTML when the editors of the source Vektor space are trusted.