OpenNG

OpenNG is an open-source REST API platform that makes Nigerian public data accessible to developers. Data that exists in government PDFs, broken portals, and Excel files is collected, cleaned, validated, and served through a versioned REST API, alongside a web app for exploration and a documentation site.

Planned production surfaces:

Surface	Role
API	Versioned JSON REST API (api.openng.dev)
Web	Landing page, data explorer, dashboard, contribution flows (openng.dev)
Docs	Product and API documentation (docs.openng.dev)

Repository: github.com/stephcrown/openng

Tech stack

Layer	Technology
Language	TypeScript (strict)
Runtime	Node.js 22+
Monorepo	pnpm workspaces, Turborepo
API	Hono on Node.js
Web	TanStack Start (Vite, TanStack Router, React)
Documentation	Next.js App Router, Fumadocs
ORM	Drizzle
Database	PostgreSQL 16 (multiple schemas: public, staging, ref, analytics)
Validation	Zod
Cache / rate limiting	Redis
CI	GitHub Actions

Shared packages use the @openng/ scope (@openng/db, @openng/shared, @openng/ui, @openng/scrapers, plus internal ESLint and TypeScript config packages).

Architecture

High-level monorepo and runtime relationships:

flowchart TB
  subgraph Users
    Browser[Browser]
    Integrators[API consumers]
  end

  subgraph Apps["apps/"]
    Web["web — TanStack Start"]
    Docs["docs — Next.js / Fumadocs"]
    API["api — Hono"]
  end

  subgraph Packages["packages/"]
    DBpkg["@openng/db — Drizzle schemas & migrations"]
    Shared["@openng/shared"]
    Scrapers["@openng/scrapers"]
    UI["@openng/ui"]
  end

  subgraph DataLayer["Data & infra"]
    PG[("PostgreSQL")]
    Redis[("Redis")]
  end

  Browser --> Web
  Browser --> Docs
  Integrators --> API
  Web --> API
  API --> PG
  API --> Redis
  DBpkg --> PG
  Scrapers --> PG
  API --> Shared
  Web --> UI
  Docs --> UI

Loading

How data moves from sources to the API (staging pipeline and production reads):

flowchart LR
  subgraph Ingestion["Ingestion"]
    Seeds[Seeds & imports]
    Jobs[Scraper jobs]
  end

  subgraph Postgres["PostgreSQL"]
    Staging[staging schema]
    Public[public schema]
    Ref[ref schema]
  end

  subgraph API["Hono API"]
    V1["/v1/* resources"]
    RefAPI["/v1/ref/*"]
  end

  Seeds --> Staging
  Jobs --> Staging
  Staging -->|"validate & migrate (stored procedures)"| Public
  Ref --> RefAPI
  Public --> V1

Loading

Getting started

Prerequisites

• Node.js 22 or newer (see root engines and .nvmrc if present)
• pnpm (workspace pins packageManager in root package.json)
• Docker for local PostgreSQL, Redis, and observability stack (docker compose)

1. Install dependencies

From the repository root:

pnpm install

2. Environment variables

Copy the example env files and fill in secrets where needed:

cp .env.example .env
cp apps/api/.env.example apps/api/.env
cp apps/web/.env.example apps/web/.env

Generate secrets as described in .env.example (for example openssl rand -hex 32 for SESSION_SECRET and MAGIC_LINK_SECRET). Root .env.example documents database URLs aligned with Docker Compose ports.

3. Start infrastructure

Postgres and Redis (and related services) run in Docker; apps run on the host in development:

docker compose up -d

4. Database migrations

With DATABASE_URL set (see root .env), apply schema migrations from @openng/db:

pnpm --filter @openng/db db:migrate

5. Run the apps

Start all packages in watch mode:

pnpm dev

Or target one app:

pnpm exec turbo dev --filter=api
pnpm exec turbo dev --filter=web
pnpm exec turbo dev --filter=docs

Default local ports:

App	Port
API (apps/api)	http://localhost:3000
Web (apps/web)	http://localhost:3001
Docs (apps/docs)	http://localhost:3002

Tests and quality

pnpm test
pnpm lint
pnpm check-types

API tests use a separate database when configured; see apps/api/.env.test.example.

Project layout (overview)

openng/
├── apps/
│   ├── api/          # Hono REST API
│   ├── web/          # TanStack Start frontend
│   └── docs/         # Fumadocs documentation site
├── packages/
│   ├── db/           # Drizzle schema and migrations
│   ├── shared/       # Shared types and helpers
│   ├── scrapers/     # Data acquisition jobs
│   └── ui/           # Shared UI primitives and tokens
├── scripts/          # Import, validate, migrate CLIs
├── data/seeds/       # Versioned seed data (where applicable)
└── docker-compose.yml

---

Contributions welcome via issues and pull requests on GitHub.