Small, free, browser-only tools for one exact developer problem each, plus a home-library app.
Live: https://arling.sk/
This repository is the source for the hub page at arling.sk, the page every ARLing tool links back to. It is one static site:
index.html: the page itself. Lists the 7 live tools, each with what you see, what you get, and a link. Has a "notify me about new tools" email form and three JSON-LD blocks (Organization,WebSite,ItemListof the 7 tools) so the company and the tool list are machine-readable.how-we-work/index.html: the method page. What Andrej decides, what AI agents do, how the company makes money, a 404 page for the section.privacy/index.html: what is collected (anonymous visit counts, an optional e-mail address) and what never is (anything typed into a tool).llms.txt: the same tool list in plain text for AI agents, with a one-paragraph summary of each tool and a link to that tool's ownllms.txt/llms-full.txt.sitemap.xmlandsitemap-index.xml: this domain's own pages, plus an index that points at every tool's ownsitemap.xml(each tool is a separate repository with its own sitemap).robots.txt:Allow: /for everyone, with GPTBot, ClaudeBot, PerplexityBot, Google-Extended, Bingbot and Applebot listed explicitly.CNAME:arling.sk. This repo is namedAndryRoby.github.io(GitHub's naming convention for a user's default Pages site) and GitHub Pages serves it at the custom domain in that file.subscribe.js: wires the email form to a self-hosted subscribe API. No inline event handlers, since the CSP inindex.htmlhas nounsafe-inlinefor scripts.google053b559de11f5b40.htmland25d148747bb8560e7423770bf741ecbe.txt: Google Search Console and IndexNow verification files.
- It runs no checks itself. Every diagnostic lives in its own tool, in its own repository, with its own engine and its own tests.
- No accounts, no login, no cookies of its own.
- It does not store what anyone types into a tool. The one thing it does store, if you submit the form, is an email address for the "new tools" list, held by the subscribe service, not in this repo.
- It has no backend of its own. The only network call this repo's code makes is the subscribe form's POST, to a self-hosted endpoint outside this repo.
index.html is one static file: inline CSS, no framework, no build step, no client-side router. The only script beyond analytics is subscribe.js, which listens for submit on form[data-subscribe], validates the email, and posts JSON to the subscribe endpoint. A 409 (already subscribed) is treated as success, not an error.
Example, from subscribe.js:
POST https://homelab.tailbf8f27.ts.net/subscribe/api/subscribe
Content-Type: application/json
{"email":"you@example.com","source":"hub","lang":"en","hp":""}
A 2xx or 409 response swaps the form for the "thanks" message; anything else swaps it for the error message, both already in index.html and just unhidden by the script.
No build step. Any static file server works:
git clone https://github.com/AndryRoby/AndryRoby.github.io.git
cd AndryRoby.github.io
python -m http.serveror open index.html directly in a browser. The subscribe form and the Umami script will fail closed (caught, logged, no crash) when the homelab endpoint is unreachable, which is expected when running offline.
There is no tests.mjs in this repository: a static index page with no engine has nothing to unit-test. Each tool it links to has its own tests.mjs, in its own repository.
Everything on this page runs in your browser except one thing: submitting the subscribe form, which is opt-in. Visit counts and which link was clicked are collected with a self-hosted Umami instance that runs without cookies; the content of anything typed into a tool linked from this page never reaches this domain, since each tool is client-side on its own page. Full policy: https://arling.sk/privacy/.
This repo has no diagnostic rules to cite, only public specs it follows for discoverability:
- schema.org
Organization/WebSite/ItemList(JSON-LD): https://schema.org/ - Sitemaps XML protocol: https://www.sitemaps.org/protocol.html
- IndexNow protocol: https://www.indexnow.org/
- robots.txt crawler directives: each bot's own published user agent (GPTBot, ClaudeBot, PerplexityBot, Google-Extended, Bingbot, Applebot).
A broken link, a wrong number on the hub page, or a tool missing from the list: open an issue on this repo, or write to andrej@arling.sk with the URL and what's wrong.
All rights reserved, ARLing s. r. o. Reading this code to see how the hub page works, or to learn from it, is fine. Deploying a copy of it as your own site is not. Each tool repository carries its own LICENSE-NOTICE.md with the exact terms; this repo follows the same policy even though the file isn't copied into it yet.
ARLing s. r. o., Bratislava, Slovakia. Company ID 56583486, VAT ID SK2122352100. andrej@arling.sk
The tools this hub links to, each its own repository: