Privacy

Last updated 31 August 2026

A tool that removes tracking has no business adding its own. This page lists everything the service keeps, in full.

What is stored

• Short links. If you tick "Also make a short link", the cleaned URL is written to a local SQLite database alongside a random six-character code and the time it was created. This is the only way /r/<code> can work.
• One number. A single integer counting how many links have been cleaned since the service started. It is not broken down by anything.

What is not stored

• Links you clean without asking for a short link. They are processed in memory and discarded when the response is sent.
• IP addresses, user agents, and referrers. The web server is configured not to write an access log.
• Cookies. The site sets none, and there is nothing to consent to.
• Analytics of any kind. No third-party script runs on this site. The fonts are served from this server, so your browser never contacts a font CDN.
• Accounts. There are none.

The one exception

Rate limiting keeps a count of requests per IP address in memory for sixty seconds at a time, so that one client cannot flood the service. Nothing is written to disk and nothing survives a restart.

Short links are public

A six-character code is short enough to guess. Anyone who tries a code that exists will be redirected to the URL behind it. Do not shorten a link you would not be comfortable handing to a stranger, and note that short links have no expiry.

The API

POST /clean is open to any origin and requires no key. It is subject to the same rules as the website: nothing about the caller is recorded, and the submitted URL is only written to disk if the request asks for a short link.

Changes

If this ever changes, the date at the top of this page changes with it, and the full history is in the public repository along with the code that backs every claim above.