BoredGame

The description states that BoredGame is a mobile-first PWA, built with Next.js App Router, deployed on Vercel, using Supabase Postgres for data storage. It uses GPT-5.6 (in multiple variants) to process shelf photos and match them against a local 126,266-row BoardGameGeek snapshot.

Key technical components include:

• Direct browser uploads to a private Supabase bucket
• Use of structured outputs from GPT-5.6-sol for shelf scanning
• A text-embedding-3-small model for search optimization
• gpt-5.6-terra for sourcing Teach Sheets and videos
• A rulebook Referee pipeline that was not functional in the submitted version due to OpenAI indexing failures

The app supports:

• Photo-based shelf scanning
• Canonical game identification
• Source-checked teaching materials (for 3 showcase games)
• Rulebook references via publisher links (no private rulebook processing)

Inference The product is a proof-of-concept or prototype, not a production-ready service.

The author states that BoredGame starts with the collection people already have, aiming to reduce friction in board-game night by eliminating cataloging, teaching, and rule disputes. The tagline — “Snap a photo of your game shelf. Play tonight.” — positions it as a tool for quick, effortless access to games.

Inference This is a self-positioned utility product focused on convenience and ease-of-use for casual board-game players.

There is no evidence of prior positioning or evolution in the description; all claims are from the author’s own account.

The description does not explicitly define a target customer segment or ideal customer profile (ICP). It implies that BoredGame targets board-game enthusiasts who want to quickly identify and play games from their personal collections.

Inference The primary user is likely someone who owns a physical board-game collection and wants an easy way to access game information without manual cataloging.

There is no evidence of pricing, monetization strategy, or business model in the description. The project is described as a hackathon submission, with no mention of revenue, subscriptions, or paid features.

Not evidenced.

The app is built using:

• Next.js App Router
• Vercel
• Supabase Postgres
• OpenAI GPT-5.6 models (gpt-5.6-sol, gpt-5.6-terra)
• TypeScript
• PostgreSQL trigram/normalized search + text embeddings

Notable technical details:

• Uses direct browser uploads to Supabase
• Implements structured outputs for reliability
• Employs server-side OpenAI calls
• Includes a rulebook Referee pipeline, but it's not functional in the current version due to indexing failures

Inference The product is technically ambitious, but the implementation shows signs of incomplete or experimental features.

The description includes:

• A real production photo that reached 26 scan candidates and 25 confirmed canonical games
• An idempotent importer that loaded 126,266 catalog rows
• Demonstrated functionality for CATAN, Carcassonne, and Horrified
• A refresh-persistent library

However:

• The rulebook Referee pipeline is not functional
• There is no evidence of user adoption or usage beyond the author’s own testing
• No revenue, customer base, or growth metrics are mentioned

Not evidenced.

The description does not mention any competitors or market context. It does not reference existing solutions for board-game cataloging or rule lookup.

Not evidenced.

1. Incomplete functionality: The rulebook Referee pipeline is non-functional due to OpenAI indexing failures.
2. No production-ready features: The app is described as a hackathon submission, not a mature product.
3. Dependency on external APIs: Heavy reliance on OpenAI and Supabase with known issues (e.g., TLS errors, HTTP 502s).
4. No monetization or user data: No evidence of revenue model or user engagement beyond the author’s own use.
5. Self-reported only: All claims are unverified, and there is no external validation.

1. What is the current status of the rulebook Referee pipeline? Is it planned for future implementation?
2. Are there any plans to monetize or scale this product beyond its current prototype state?
3. How does the app handle edge cases in shelf scanning (e.g., overlapping games, poor lighting)?
4. Has the team considered integrating with existing board-game platforms or APIs beyond BoardGameGeek?
5. What are the technical and financial risks of relying on OpenAI’s API for core functionality?