MoodMap — Community Health Dashboard

Real-time toxicity monitoring for Reddit mods. Auto-scores every comment, shows a live health score, trend chart, and instant alerts — all in one dashboard.

Comment

Inspiration

Reddit moderators have no real-time way to gauge their community's health. They manually read through hundreds of comments hoping to catch toxic behavior before it escalates. I wanted to fix that with a single dashboard that tells mods everything they need to know at a glance.

What it does

MoodMap is a real-time community health dashboard for Reddit moderators. It automatically scores every comment using a multi-tier keyword engine, then displays:

  • A live Community Health Score (0-100)
  • A 24-hour toxicity trend chart
  • Recent toxic alerts with comment excerpts and direct "View in subreddit" links
  • A Top Offenders leaderboard showing the most toxic users
  • Per-user toxicity tracking stored in Redis

All accessible in one click from the subreddit's mod menu.

How I built it

Built with TypeScript, Devvit SDK (React web template), Hono server, and Redis. A comment trigger fires on every new comment and stores it in Redis. The keyword scorer processes comments on dashboard load using a multi-tier system with 50+ phrases, severity weighting, ALL CAPS detection, and spam signals. No external APIs needed — everything runs on Reddit's infrastructure.

Challenges I ran into

Devvit blocks external HTTP calls so I couldn't use third-party moderation APIs. I built a custom keyword scorer entirely from scratch with five severity tiers. Also learned that .env files don't work on Reddit servers and Redis has different methods than standard implementations.

Accomplishments that I'm proud of

Built a fully working mod tool from scratch as a beginner coder with basic Python/JS background. The dashboard looks and feels like a real production app, not a student project.

What I learned

How to build real production apps on Devvit, Redis data patterns for time-series data, and how to design mod tools that solve real moderator pain points.

What's next for MoodMap

  • Real ML-powered scoring using AI moderation APIs for far more accurate toxicity detection beyond keyword matching
  • Configurable toxicity thresholds per subreddit set by mods
  • Auto-moderation actions (auto-remove or auto-flag comments above threshold)
  • Weekly community health report posts generated automatically
  • Multi-subreddit support for mod teams managing multiple communities
  • Sentiment analysis to detect passive-aggressive and sarcastic toxic behavior that keywords miss

Built With

Share this project:

Updates