Follow the Tagr bot to filter out NSFW and harassing text. Tagr bot screens the text people followed by Nos users for NSFW, illegal, and harassing content and displays a Tagr overlay of the content for those who follow the bot in apps that support content labeling such as Nos. Users who follow the bot still have the option to view the content if they choose.

Steps to use the bot:
1. Subscribe to relay.nos.social
2. Follow the tagr-bot
3. Use an app like Nos that supports note labeling.

Last Notes

FAQ
Is anyone manually reviewing the bot?
- Yes we are monitoring the bot and making adjustments as we go
What if my content is flagged but the bot missed the context?
- reach out via email to [email protected] with the event ID of the note in question and we’ll review.
Does the bot crawl all of Nostr?
- No. Tagr only reviews the notes of people followed by Nos users.
Why do I receive a notification when my content is tagged?
- Some clients think all events should be notified.
- Nos does not believe that is necessary, especially in cases of harassment.

2024-06-03 17:11:31 GMT

How does the Tagr bot work?
The bot reviews all the notes of people Nos users follow and evaluates them for all the categories in the Open AI content moderation. We then map those to NIP69 which are then mapped to NIP56 and we display them as the following 4 types labels:
- Spam
- Sexual & Nudity
- Antisocial behavior
- Illegal
If it finds a match it labels the content. Those who follow Tagr and use a client that supports content labels will see a warning over the note.

2024-05-28 16:18:51 GMT

Tagr bot
Nostr is an open community, but some users may prefer not to see certain types of content. Platform social media has struggled for years to moderate content in a centralized system.
By putting users in control of who they follow and by extension what they see, Nostr offers another option.
However, manual content moderation can be time consuming and mentally taxing. People come to social media to connect not to serve as moderators of content in their feed.
Following the Tagr bot allows users to offload this work to AI for all text based content. AI moderation alone is not sufficient to create a healthy social network, but we think it can be part of the solution so we are using this bot to learn and experiment.

2024-05-23 15:03:40 GMT

How does the Tagr bot work?
The bot reviews all the notes in which a user is mentioned or replied to and evaluates them for all the categories in the Open AI content moderation. We then map those to NIP69 which are then mapped to NIP56 and we display them as the following 4 types labels:
- Spam
- Sexual & Nudity
- Antisocial behavior
- Illegal
If it finds a match it labels the content. Those who follow Nos and use a client that supports content labels will see a warning over the note.