Skip the legal document — here’s the plain-language version of what SpecShot collects, in one page. The legally binding detail is in the Privacy Policy.
Never leave your device. There is no upload button that sends them anywhere, because there is no server for them to go to.
None exists — there's no sign-up, no login, no password, no email required to use any tool.
One: an anonymous visit counter (Vercel Web Analytics) that sets no cookie and can't identify you — just aggregate traffic numbers. No pixel, no session recorder, no “how did you use this page” script.
Not built. No accounts exist, and the visit counter above forgets you within 24 hours by design — there's nothing that links one visit to the next.
None from SpecShot itself. Google AdSense may set one — only if you accept the banner on your first visit.
Two small, functional things: a same-tab handoff when you move an image between tools (deleted the instant it's read), and your ad-cookie choice. Neither is ever sent anywhere — see the table below.
Most sites that touch photos have to upload them somewhere to process them. SpecShot doesn’t, because every operation — face detection, background removal, cropping, compression, format conversion, watermarking, HEIC decoding, PDF reading and writing, signature cleanup — runs as WebAssembly or Canvas code that ships to your browser and executes there. The file never becomes a request your device sends out; it stays a local variable in a tab you can close at any time to erase it.
| What | Where it's stored | Ever transmitted? | Gone when |
|---|---|---|---|
| Your photo / file mid-edit | Browser memory (RAM) only | Never | You close or refresh the tab |
| Cross-tool image handoff | Session storage | Never | Read once, or tab closed |
| Ad-cookie choice | Local storage | Never | You reset it, or clear browser data |
| App code / ML model files | Service-worker cache | N/A — this is the code, not your data | Superseded by the next release, or cleared by you |