About the Project: OmniBot Mod Concierge OmniBot is a next-generation moderation suite built natively on Reddit's Devvit platform. It was designed to replace the often punitive, confusing experience of traditional AutoModerator setups with a helpful, automated assistant that guides users to follow community rules in real-time. By prioritizing user education over immediate post removal, OmniBot creates a welcoming environment while reducing the manual workload for moderator teams to zero.
Core Features & Technical Implementation:
Stateful Context Gatekeeper (Redis Integration): Many communities require Original Posters (OP) to provide a source or background context in the comments. OmniBot automates this by leaving a pinned comment requesting context and storing the Post ID in Devvit's Redis database. It then listens to the CommentSubmit trigger, verifies when the OP has replied, updates the state, and leaves a confirmation message.
Intelligent Auto-Flairing: The app listens to the PostSubmit event, scans the post title for specific keyword mapping (e.g., words like "fix" or "issue"), and utilizes the Reddit API to instantly assign the correct Post Flair (e.g., "Support"), entirely bypassing the need for users to manually categorize their posts.
Gentle Title Enforcement: Instead of ruthlessly deleting posts that lack required formatting tags (such as [Help] or [Discussion]), OmniBot scans the regex of the title. If a tag is missing, it dynamically injects a polite formatting reminder into its pinned concierge comment, educating the user for their next post without disrupting their current one.
The Impact: OmniBot transforms chaotic subreddits into self-managing systems. It allows moderators to step back from tedious chores like flairing posts or chasing users for sources, and redirects their focus toward thoughtful community building and high-level engagement.
InspirationBuilding automation tools and comprehensive applications has always been a core focus for our team. From developing robust institutional software to tackling complex problems at various hackathons, we have seen firsthand that punitive, heavy-handed systems frustrate users and stifle engagement. We were inspired to build OmniBot because we believe moderation shouldn't just be about silently deleting rule-breaking content—it should act as a concierge service that educates users and guides them to contribute meaningfully. We wanted to replace the cold, stateless nature of AutoModerator with an interactive, stateful assistant.How We Built ItWe built OmniBot entirely on Reddit's Devvit platform using TypeScript and Node.js.Event Architecture: We utilized the PostSubmit and CommentSubmit triggers to intercept user actions in real-time.State Management: We leveraged Devvit's built-in Redis integration to power the "Context Gatekeeper." When a post is submitted, we store the Post ID in Redis as a pending_context state and await the Original Poster's comment to clear it.Dynamic API Integration: We mapped keyword arrays to the Reddit API to dynamically assign Post Flairs, and we used Regex pattern matching on the post titles to construct dynamic, customized warning messages in a single pinned comment.Challenges We FacedThe Flair API: Applying flairs programmatically wasn't as simple as passing a text string. We had to navigate how the Devvit API handles specific flairTemplateId structures to ensure the visual flairs actually attached to the frontend UI.Stateful Logic: Moving from a traditional, stateless moderation mindset to a multi-step, stateful workflow required careful tracking. We had to build strict checks to ensure the bot only responded to the actual OP of a thread and didn't allow random users to trigger the Gatekeeper approval process.The Deadline Sprint: Executing a complete architectural pivot and pushing a fully functional, bug-free prototype on the final day of the hackathon tested our rapid-prototyping skills to the absolute limit!What We LearnedTransitioning into the Devvit ecosystem deepened our understanding of event-driven architectures within a tightly scoped platform. We also learned how to clearly define and quantify the impact of UX design on moderation.We can actually model the exact efficiency OmniBot brings to a community. If $t$ represents the average time a moderator takes to manually verify a post's context, $N$ is the total daily submissions, and $p$ is the probability of a formatting violation, the total manual time spent by a human mod team is:$$T_{manual} = \sum_{i=1}^{N} (p_i \cdot t_i)$$By shifting this workload to OmniBot's automated event triggers, this equation approaches zero. The bot handles the algorithmic enforcement, freeing up the human moderators to focus entirely on organic community building.
Log in or sign up for Devpost to join the conversation.