Working software, kept deliberately small. Each project is built to solve or explore a specific problem and, where appropriate, released openly so the implementation can be inspected.
A small, local-first website crawler built with Crawlee and Playwright.
Give it a website URL and a page limit. Kalax Crawl follows same-site pages, shows crawl activity as it runs, and returns each page's URL, title and HTTP status. Results can be downloaded as CSV or JSON.
Clone it and run it on your own machine.
No external API credential is required to crawl.
The V1 product is deliberately local-first.
A real local execution: starting a crawl, watching live progress, and reaching the completed result.
Kalax Crawl is published as working source code. Clone the repository, install it locally and run the crawler yourself.
View Kalax Crawl on GitHub (opens in a new tab)What does your website actually tell the world?
SiteCheck examines what customers, search systems and machines can actually understand from a website, then shows the evidence behind each finding.
Can people find a clear way to contact or enquire with you?
Can search systems reach, identify and navigate the pages checked?
Does the site expose a consistent machine-readable identity?
Can machines parse the site structure and supporting information?
Findings use explicit evidence and preserve what could not be verified instead of turning uncertainty into a score.
Download a readable PDF report or the underlying JSON evidence from the completed check.
Runs locally with no account, API key, database or paid service required for V1.