Discussion
Discord open sourced their internal trust and safety rules engine with @roost and has now reintegrated it into production!
Osprey now handles around 400 million actions per day in production at Discord. If you run a Discord-sized platform (or anything smaller!), you can just… use their automated rules engine yourself because it’s open source.
https://discord.com/blog/osprey-open-sourcing-our-rule-engine
@matrix is doing just that by integrating Osprey into their policy servers for automated rules and investigations. Users of the open source, decentralized alternative platform will be safer, too, thanks in part to @roost. 😉
Bluesky is also already using Osprey to handle tens of millions of daily events, tackling real-time social media issues like scams, spam, and more.
https://roost.tools/blog/how-bluesky-transformed-their-safety-operations-with-osprey/
It’s an exciting time for open source trust and safety!
If Osprey is overkill (it can be for small/early platforms!), we also just open sourced Coop, a more straightforward trust and safety review dashboard with its own automation built in—including optional integrations with hash matching, Google Content Safety API, and OpenAI moderation API.
https://roost.tools/blog/meet-coop-v-0/
While Osprey’s actively used in multiple production environments, Coop is still v0; it’s out there but we’re actively working towards a stable release.
@cassidy
> If Osprey is overkill (it can be for small/early platforms!), we also just open sourced Coop, a more straightforward trust and safety review dashboard with its own automation built
