Great stuff!

I don't know of a way a single person can prove they haven't changed the webpage content they archived. In other words, a person can easily first change the webpage content and then generate the hashes. So others have to trust that the person who archived didn't make any changes before archiving.

To overcome this, there can be verifiers who very quickly check a URL right after it gets archived and verify that the archived content matches the content on the URL at this moment in time. I'm not sure we can get around the need for verifiers. But the use case seems solid and it should work. And I really like the client-side and bookmarklets approach. :)