Medusa Auto-Mod

Inspiration

Reddit moderators spend hours manually removing spam and toxic content. Medusa automates this so mods can focus on community building.

What it does

Medusa Auto-Mod monitors every post and comment in real-time. When spam, toxic words, or forbidden links are detected, Medusa automatically removes the content and posts a warning comment explaining the violation.

How we built it

Built on Reddit's Devvit platform using TypeScript. Used onCommentSubmit and onPostSubmit triggers to intercept content, pattern matching for detection, and reddit.remove() for enforcement.

Challenges we ran into

Devvit's architecture required careful loop-guard implementation to prevent Medusa from analyzing her own warning comments and creating infinite trigger loops.

Accomplishments that we're proud of

Real-time moderation working end-to-end — spam posted, spam removed, warning posted, all within seconds.

What we learned

Devvit's trigger system is powerful but requires precise permission scoping in devvit.json.

What's next for Medusa Auto-Mod

Claude API integration for semantic analysis beyond pattern matching.

Built With

• devvit
• redditapi
• typescript