There is an update for nostr-filter-relay documentation and explanation on "how it works".
What do you think? Is it clear explanation or still vague?

GitHub
GitHub - atrifat/nostr-filter-relay: A Nostr relay docker image package which filter content based on content type (SFW/NSFW), user type, language, hate speech (toxic comment), sentiment, topic, and various rules.
A Nostr relay docker image package which filter content based on content type (SFW/NSFW), user type, language, hate speech (toxic comment), sentime...

The only downside is it doesn't seems support url parameters yet.
Nostrudel on the other hand has already support url parameters
And if we like simplicity, nostrrr.com and njump.me are good choices because they only give text without rendering many graphics.
New feature of nostr-filter-relay (hate speech/toxic comment detection) has been available in dev branch of code and in test-relay instance. It will be released soon after some testing.
If you want to see some toxic contents ๐
:
wss://nostr-id-test-relay.hf.space/?toxic=yes&toxic_confidence=75