MidNight Network

A ranking aggregation and leaderboard system for a daily scheduled post-timing competition game on Misskey

Overview

MidNight Network is a comprehensive web application for a game where participants join by posting notes on Misskey at a scheduled daily time. When a user posts a specific string, the system aggregates the post timestamps to rank and visualize results.

Key Features

7-Tier (13 with intermediate) Ranking System — Point-based progression from unranked to the highest tier
Automated Match Processing — Daily results aggregated from Misskey timeline via cron jobs
Multiple Leaderboards — Rank points, win rate, average time, and match time
User Profiles & Analytics — Detailed statistics with charts (earned points, heatmaps, radar, post time trends)
Streak Tracking — Consecutive participation, within-zone, absence, and flying streaks with bonus points
Border Protection — Demotion prevention system for players near rank boundaries
Push Notifications — Match results delivered via Web Push API
Privacy Controls — Granular visibility settings for leaderboards and profiles

Tech Stack

Layer	Technologies
Frontend	SvelteKit (Svelte 5), Vite, TanStack Query, TanStack Table, ECharts, Motion
Backend	Cloudflare Workers, Elysia, oRPC
Auth	Better Auth + MiAuth (Misskey)
Database	Cloudflare D1 (SQLite) + Prisma ORM
Shared	oRPC contracts, Zod schema validation
Rank Engine	Custom TypeScript calculation module
Code Quality	Biome (lint + format), Lefthook (git hooks)

Project Structure

This is a pnpm workspace monorepo:

packages/
├── backend/     Cloudflare Workers API server (Elysia + Prisma + D1)
├── frontend/    SvelteKit web application (static adapter)
├── shared/      oRPC contracts & shared types
└── rank-calc/   Ranking calculation engine

Prerequisites

Node.js: 24^
pnpm: 10^
Cloudflare: Account (for D1, KV, R2, Workers, Queues)
Misskey: Instance access and API token

Getting Started

1. Clone the repository

git clone https://github.com/Secineralyr/midnight-network.git
cd midnight-network

2. Install dependencies

pnpm install --frozen-lockfile

3. Configure environment variables

Copy the backend template and fill in the required values:

cp packages/backend/.dev.vars.template packages/backend/.dev.vars

Variable	Description
`MK_HOST`	Misskey instance hostname
`API_TOKEN`	Misskey API token
`WEB_HOST`	Frontend hostname
`BACKEND_HOST`	Backend hostname
`BOT_USER_NAME`	Bot user name on Misskey
`DISABLE_USER_NAME`	User IDs to exclude (JSON array `string[]`)
`ADMIN_USER_NAME`	Admin user name
`WEBHOOK_SECRET`	Webhook secret key
`VAPID_PRIVATE_KEY`	VAPID private key for push notifications
`BETTER_AUTH_SECRET`	Secret key for auth session encryption (32+ chars)

Frontend — Copy the template and configure:

cp packages/frontend/template.env packages/frontend/.env

Variable	Description
`VITE_API_ROOT`	Backend API URL (empty for same-origin)
`VITE_MOCK_AUTH`	Enable mock auth for development
`VITE_TARGET_HOUR`	Match target hour in UTC (default: `15`)
`VITE_TARGET_MINUTES`	Match target minutes in UTC (default: `0`)
`VITE_VAPID_PUBLIC_KEY`	VAPID public key for push notifications

4. Local database migration

This step is required even if you are not developing, as it generates the Prisma types.

pnpm --filter @midnight-network/backend migrate:local

5. Start development servers (for development)

# Frontend (port 5173)
pnpm front:dev

# Backend (port 8787)
pnpm back:dev

The frontend dev server automatically proxies /api requests to the backend.

6. Build (for deployment)

pnpm -r build

7. Backend `wrangler.toml` configuration

The following items need to be defined in wrangler. It is recommended to split environments for the modified items.

The following items are primarily required for deployment.

routes: The backend allows api and webhook paths for the host
kv_namespaces: Replace id with the ID generated by pnpm wrangler kv namespace create <name> (note that two are required: SESSION and CACHE!)
r2_buckets: Replace bucket_name with the bucket name generated by pnpm wrangler r2 bucket create <bucket_name>
queues (producers and consumers): Replace queue with the name generated by pnpm wrangler queues create <name> (note that producers and consumers must use the same queue!)
d1_databases: Replace database_name and database_id with those generated by pnpm wrangler d1 create <name>

Additionally, edit the following items as needed:

Variable	Description
`ENVIRONMENT`	Current environment; do not set to production outside of the production environment
`GAME_JOIN_POST_TEXT_REGEX`	Regex pattern for matching note text to aggregate
`TARGET_MATCH_HOUR`	Target aggregation hour in UTC
`TARGET_MATCH_MINUTES`	Target aggregation minutes in UTC
`POST_MATCH_RESULT_TITLE`	Title used when posting aggregation results as a note
`POST_MATCH_RESULT_URL`	Frontend URL included in the result note
`POST_MATCH_RESULT_TEMPLATE`	Result note text template (use curly braces for format placeholders: `{title}`: title / `{ranks}`: ranking (top 10) / `{valid}`: valid record count / `{flying}`: flying record count / `{url}`: URL)
`DISABLE_POST_MATCH_RESULT`	Disable posting result notes after aggregation
`POST_MATCH_REMIND_TEXT`	Reminder note text posted before the match
`DISABLE_WEBHOOK_BEFORE_TARGET_MATCH`	How long before the aggregation time to ignore webhook responses (ms)
`DISABLE_WEBHOOK_AFTER_TARGET_MATCH`	How long after the aggregation time to re-enable webhook responses (ms)
`VAPID_PUBLIC_KEY`	VAPID public key for push notifications
`DB_QUERY_LOGGING`	Log the executed query (for debug)

There are also cron execution settings. These exist because triggers.cron alone cannot control individual processes, so these definitions are used for that control. Therefore, you must assign the cron trigger string you want to execute to the corresponding process.

Variable	Description
`CRON_DAILY`	Aggregation process
`CRON_DAILY_REMIND`	Reminder post process

8. Frontend `wrangler.toml` configuration

When deploying the frontend with Cloudflare Workers, the following items need to be defined in wrangler. It is recommended to split environments for the modified items.

routes: Allow any path for the host

Dev Container (VS Code)

A Dev Container configuration is included for VS Code:

Install the Dev Containers extension
Open the project and select "Reopen in Container"
Dependencies are installed automatically on first launch
Configure packages/backend/.dev.vars as described above
Start development with pnpm front:dev / pnpm back:dev

Forwarded Ports:

Port	Service
`5173`	Frontend (SvelteKit dev server)
`8787`	Backend (Wrangler dev server)

Available Scripts

Root (all packages)

Command	Description
`pnpm install`	Install all dependencies
`pnpm check`	Run Biome lint + format check
`pnpm check:write`	Auto-fix lint and format issues
`pnpm typecheck`	Type-check all packages
`pnpm build`	Build all packages
`pnpm test`	Run all tests
`pnpm front:dev`	Start frontend dev server
`pnpm back:dev`	Start backend dev server

Backend

Command	Description
`pnpm --filter @midnight-network/backend dev`	Start Wrangler dev server
`pnpm --filter @midnight-network/backend build`	Build (generate types + compile)
`pnpm --filter @midnight-network/backend migrate:local`	Apply D1 migrations locally
`pnpm --filter @midnight-network/backend migrate:remote`	Apply D1 migrations to production
`pnpm --filter @midnight-network/backend deploy`	Deploy to Cloudflare Workers
`pnpm --filter @midnight-network/backend auth:gen`	Update Prisma definitions when Better Auth definitions change
`pnpm --filter @midnight-network/backend prisma:format`	Format Prisma files
`pnpm --filter @midnight-network/backend prisma:mig`	Write diff from Prisma definitions to migration file (may be inaccurate; manually defining is generally recommended)

Frontend

Command	Description
`pnpm --filter @midnight-network/frontend dev`	Start Vite dev server
`pnpm --filter @midnight-network/frontend build`	Build static site
`pnpm --filter @midnight-network/frontend test`	Run Playwright E2E tests

Deployment

CI/CD (GitHub Actions)

Automated CI and deployment workflows are configured in .github/workflows/.

CI

Runs on pull requests:

Lint & format check (Biome)
Local D1 migration (for Prisma type generation)
Build all packages
Typecheck
Tests

Deploy

Runs automatically on push:

Builds shared packages, backend, and frontend
Replaces wrangler.toml with the production/staging version stored in GitHub Secrets
Deploys backend and frontend to Cloudflare Workers via wrangler deploy
Runs remote D1 migrations after backend deployment
Frontend VITE_* environment variables are injected at build time

Required GitHub Secrets

Secret	Description
`CLOUDFLARE_API_TOKEN`	Wrangler API token (Workers edit permission)
`CLOUDFLARE_ACCOUNT_ID`	Cloudflare account ID
`BACKEND_WRANGLER_TOML`	Full contents of the backend `wrangler.toml` for this environment
`FRONTEND_WRANGLER_TOML`	Full contents of the frontend `wrangler.toml` for this environment
`VITE_API_ROOT`	Frontend API root (empty for same-origin)
`VAPID_PUBLIC_KEY`	VAPID public key for push notifications
`D1_DB_NAME`	D1 database name

Note: Worker Secrets (e.g. API_TOKEN, WEBHOOK_SECRET, BETTER_AUTH_SECRET) must be configured directly on Cloudflare (via Dashboard or wrangler secret put). They are not managed by the deployment workflow.

Manual Deployment

Backend

Deploy as a Cloudflare Worker

pnpm --filter @midnight-network/backend deploy
# Using environment-based deployment is recommended
# pnpm --filter @midnight-network/backend deploy --env <env>

Required Cloudflare resources:

D1 Database: Main data store
KV Namespaces: SESSION (auth sessions), CACHE (general cache)
R2 Bucket: Rank status snapshots
Queue: Rerun job processing

Configure resource bindings in packages/backend/wrangler.toml.

Frontend

The frontend is built as a static site using @sveltejs/adapter-static:

pnpm --filter @midnight-network/frontend build

Output is generated in packages/frontend/dist/ and can be deployed to any static hosting service (Cloudflare Workers, Vercel, Netlify, etc.). Deploying to Cloudflare Workers is generally recommended.

pnpm --filter @midnight-network/frontend deploy
# Using environment-based deployment is recommended
# pnpm --filter @midnight-network/frontend deploy --env <env>

Contributing

Contributions are welcome! Please see CONTRIBUTING.md for guidelines.

License

The source code of this project is licensed under AGPL-3.0-or-later. See LICENSE for details.

However, images and assets (excluding manifest files) in the packages/frontend/static/ directory are not open source and are provided under a proprietary license. See packages/frontend/static/LICENSE for details.

If you fork this project, you must replace the proprietary assets with your own.

Name		Name	Last commit message	Last commit date
.devcontainer		.devcontainer
.github		.github
.vscode		.vscode
LICENSES		LICENSES
packages		packages
.editorconfig		.editorconfig
.gitignore		.gitignore
.node-version		.node-version
.npmrc		.npmrc
CODE_OF_CONDUCT.md		CODE_OF_CONDUCT.md
CONTRIBUTING.md		CONTRIBUTING.md
LICENSE		LICENSE
README-jp.md		README-jp.md
README.md		README.md
REUSE.toml		REUSE.toml
biome.json		biome.json
lefthook.yml		lefthook.yml
package.json		package.json
pnpm-lock.yaml		pnpm-lock.yaml
pnpm-workspace.yaml		pnpm-workspace.yaml
tsconfig.base.json		tsconfig.base.json

Navigation Menu

Uh oh!

Repository files navigation

MidNight Network

Overview

Key Features

Tech Stack

Project Structure

Prerequisites

Getting Started

1. Clone the repository

2. Install dependencies

3. Configure environment variables

4. Local database migration

5. Start development servers (for development)

6. Build (for deployment)

7. Backend `wrangler.toml` configuration

8. Frontend `wrangler.toml` configuration

Dev Container (VS Code)

Available Scripts

Root (all packages)

Backend

Frontend

Deployment

CI/CD (GitHub Actions)

CI

Deploy

Required GitHub Secrets

Manual Deployment

Backend

Frontend

Contributing

License

About

Contributors

Languages

Uh oh!

Latest commit

History

Folders and files

Repository files navigation

MidNight Network

Overview

Key Features

Tech Stack

Project Structure

Prerequisites

Getting Started

1. Clone the repository

2. Install dependencies

3. Configure environment variables

4. Local database migration

5. Start development servers (for development)

6. Build (for deployment)

7. Backend wrangler.toml configuration

8. Frontend wrangler.toml configuration

Dev Container (VS Code)

Available Scripts

Root (all packages)

Backend

Frontend

Deployment

CI/CD (GitHub Actions)

CI

Deploy

Required GitHub Secrets

Manual Deployment

Backend

Frontend

Contributing

License

About

Resources

Code of conduct

Contributing

Stars

Watchers

Forks

Contributors

Languages

7. Backend `wrangler.toml` configuration

8. Frontend `wrangler.toml` configuration