What we collect (almost nothing)

Files you convert are never uploaded to our servers. The entire conversion runs in your browser using JavaScript and WebAssembly. We don't have servers handling files because there's no server-side step in any conversion route.

This means we cannot see, log, store, or recover what you convert , even if we wanted to. The same conversion library that runs on CloudConvert's servers (FFmpeg, libheif, pdfjs, mammoth, etc.) runs on your machine instead.

We keep standard server access logs (IP address, page URL, timestamp, browser user-agent) for the static pages you visit. These logs are rotated automatically and used only for debugging deployment issues and counting page views per route.

We do not use Google Analytics, Mixpanel, Heap, FullStory, or any session-recording service. There is no individual-level tracking across pages or sessions.

twineconvert sets no cookies of its own. We do not have accounts, login state, or personalization that would require cookies.

When advertising launches on the site (which is the eventual revenue model), the ad provider, Google AdSense initially, Mediavine or Raptive once we hit their traffic threshold, will set their own cookies for ad measurement. Those cookies are governed by the provider's privacy policy, and we'll update this page when ads go live with a clear opt-out path.

The site loads its conversion libraries (FFmpeg.wasm core, pdfjs worker, web-ifc WASM) from a public CDN (unpkg.com) the first time you use a tool that needs them. The CDN sees your IP address as part of any normal HTTP request. After the first load these files are cached in your browser and served from disk.

The site is hosted on Vercel; Vercel sees the same standard server access logs described above.

Because we don't collect personal data, there's nothing for you to request, delete, or correct. If you have questions, the GitHub repository at github.com/Achraf921/conversionEngine is the project's home, you can read every line of code that handles a file, and open an issue if something looks off.

If we ever add server-side anything (paid tiers, accounts, file history), this page will be updated before the change ships, not after. The current commit history of this file lives in the open-source repo, so any changes are public-record.