Inspiration
As a moderator of a busy community, I was constantly fighting "dogpiling." Bad actors know that native AutoModerator relies on aggregate report volume. To force a removal, they submit a scattershot of different, inaccurate reports just to hit that global threshold and so this wasn't suitable. At the same time, rule-breaking content was staying up for extended periods when no moderators were online, even when it was reported by users who were available. I needed a tool that evaluates true consensus on a specific rule violation and actions removals.
How it was built
I built the app using Devvit to listen natively for PostReport and CommentReport events. When triggered, the bot queries the Reddit API for the live target object, parses the user reports, and cross-references the grouped report counts against custom thresholds configured for the subreddit before filtering the post.
Challenges faced
The biggest hurdle was handling bad-faith reporting safely. Initially, I envisioned a hard remove action. However, realizing this could be weaponized by trolls, I pivoted the core logic to filter. This ensures actioned items are pulled from the public feed immediately but remain in the modqueue for final human review.
What I learned
I learned the critical difference between a trigger's initial event payload and the necessity of fetching the live state via the Reddit API to accurately calculate grouped report metrics.
Log in or sign up for Devpost to join the conversation.