Building Forgeboard for DOGFOOD 2026: a hackathon judging platform that refuses to fake confidence

Description and use

Forgeboard is an offline‑ready, single‑container hackathon judging platform that enforces strict access control and data integrity using SQLite and Next.js.

Problem

Hackathon organizers needed a self‑hostable judging platform that could run offline, enforce strict access controls, and handle complex data fixtures without data corruption.

Outcome

Forgeboard meets DOGFOOD’s strict offline and data‑integrity requirements, passes all official checks, and demonstrates robust judge isolation and data validation.

Workflow

• 1. Run docker compose up --network none to start the container.
• 2. The app auto‑seeds from fixtures.json, logging dropped data such as duplicate submissions.
• 3. Organizers configure events, tracks, prizes, and rubrics via the web UI.
• 4. Teams join via invite links, draft and submit projects before deadlines.
• 5. Judges receive scoped invitations, view only their assigned projects, and score using weighted rubrics.
• 6. The system normalizes scores across judges and publishes results to a public gallery.

Evidence

One Next.js app, one SQLite file, one container. docker compose up gives you a seeded portal, and it keeps working with the network switched off.

Source

The official DOGFOOD checker passes 7 of 7. It only has checks for T1 and T2, so T3 and T4 are covered by our own evidence: 131 unit and integration tests, 10 live checks against the running app, and browser suites that drive real Chrome.

Source

Rules live in the data layer. On every request, Forgeboard builds a capability, meaning what one actor may do inside one event, from the event_roles table.

Source