The pitch: metadata is gone, but the source can still be recovered.

This page is a controlled test library for ContentSeal. Each image has a specific outcome, so teammates and reviewers can see exact-source proof, screenshot recovery, modified-copy warning, proof deletion, and honest unknown handling.

Start by Sealing Source Recover Source

Verification cases

Fixture gallery

These are test inputs, not decorative artwork. Use them to prove the system can recover source evidence after screenshot/repost damage.

scanning fixture 01

analysis progress0%

Verified Original

Official Event Poster

Baseline for the product promise: the original file is sealed before it spreads, creating a trusted source that later scans can recover.

file

demo/assets/01-original-proof.png

action

Seal this file first, then recover it from the same upload

expect

Exact SHA-256 match, active issuer proof, proof page available

Recover this case Seal source first

Test data library

Where the demo data lives

The live demo includes public previews, while local testing can upload the files from demo/assets. Regenerating fixtures rewrites both upload-ready files and page previews.

local upload: demo/assets/*.png

preview: public/demo/assets/*.png

regenerate: npm.cmd run demo:assets

Demo closing line

Fake detection asks users to guess after content spreads. ContentSeal recovers trusted source evidence after screenshots, reposts, compression, and crops break the metadata trail.

See Trust Engine