Create a static page

Frugal configuration

First you need to write some configuration module for frugal. In a module (the convention is /frugal.config.ts, but you can use any other path) export a simple config object:

import type * as frugal from '';

const self = new URL(import.meta.url);

export const config: frugal.Config = {
    outputDir: './dist',

The self value will be used to define the root of your project. Since it is conventional to have the configuration at the root of the project, self should be the absolute url of the module. Every relative path in the config will be resolved relatively to the root of your project.

The outputDir value is the path where Frugal will generate your site. Inside this directory Frugal will create a public directory that can be served by a server like nginx or Apache HTTP Server.

Static page descriptor

In a module (/pages/posts/detail.ts for example), create a static page descriptor:

import type * as frugal from '';
import { queryPost, queryPostList } from './api.ts';
import { post } from './component/post.ts';

export const pattern = '/post/:id';

type Path = { id: string };

export async function getPathList(): Promise<Path[]> {
    const postList = await queryPostList();
    return => ({

export async function getStaticData(
    context: GetStaticDataContext<Path>,
): Promise<DataResult<Post>> {
    const post = await queryPost(;
    return { data: post };

export function getContent(params: frugal.GetContentParams<Path, Post>) {
    return `<html>

export const self = new URL(import.meta.url);

The pattern value defines the route of your page. You can use url parameters with path-to-regexp syntax.

The getPathList method should return a list of parameter object that will be used to generate a list of path with pattern. If you use a headless CMS for example, this is typically where you'll fetch the list of entities you want to generate pages for. This function can be async.

The getStaticData method will be called for each path returned by the getPathList and should return a data object that will be consumed to produce the view. Everything you need from a distant source (database, file, api) to generate your page, you should fetch here. This function can be async

The getContent method will be called with each data object returned by getStaticData, and should return the full html of your page. This function can be async.

The self constant should be the absolute url of the module. Unless you really know what you are doing, it should always be new URL(import.meta.url).

After creating the page descriptor, we need to register it in /frugal.config.ts:

import * as postDetail from './pages/posts/detail.ts';

export const config: frugal.Config = {
    pages: [,

The pages array contains all page descriptors you wish to include in your site.

Build the website

With only static pages, Frugal will effectively be a static site generator (SSG), with only a build step. In a module /build.ts add the following code:

import { config } from './frugal.config.ts';
import { build } from '';

await build(config);

Running this module will generates all pages in the /public directory. Frugal build is incremental. This means that Frugal will skip building a page if both those conditions are true:

  • the code of the page (and its dependencies) did not change
  • the data of the page did not change

This means that Frugal only ever do the minimal amount of work needed to update your site. If you change only 3 pages, Frugal will only build those 3 pages.

You should fetch all your data in the getStaticData function and never in getContent. If you fetch in getContent, Frugal will never be able to know if the data has changed and needs to be rebuilt.