OpenSyria Website

Public website for OpenSyria, a public data commons for reliable Syrian datasets, API access, and civic intelligence.

The website is intentionally small: a localized landing page, FAQ content, SEO metadata, social images, contributor attribution, theme and language controls, and links into the public API documentation and GitHub organization.

Public URLs

URL	Purpose
https://opensyria.org	Public website
https://api.opensyria.org/docs	API documentation
https://github.com/Open-Syria	GitHub organization

Agent Discovery

The site publishes public, read-only discovery metadata for agents:

• /llms.txt and /index.md describe the project and link to the main public resources.
• /auth.md explains that public website and dataset API access does not require registration, OAuth, API keys, or credentials.
• /.well-known/api-catalog links to the public API documentation, OpenAPI description, health endpoint, and dataset pages.
• /.well-known/agent-skills/index.json lists the available OpenSyria agent skills.
• OAuth/OIDC and MCP well-known routes return explicit 404 application/problem+json responses until OpenSyria offers protected auth flows or a public MCP server. Both /.well-known/mcp/server-card.json and the scanner-compatible plural alias /.well-known/mcp/server-cards.json use that unsupported response.

Stack

• Next.js 16 App Router with Cache Components enabled
• React 19
• next-intl with localePrefix: "as-needed"
• shadcn Base UI components
• Tailwind CSS 4
• Biome for formatting and linting
• pnpm 11 with supply-chain protections
• Docker standalone Next.js runtime
• GitHub Actions deployment to opensyria-prod

Repository Layout

src/app/[locale]/        Localized app routes and metadata
src/components/          Website UI components
src/components/ui/       shadcn/Base UI primitives
src/i18n/                next-intl routing, navigation, and request config
src/lib/                 Site config and GitHub data helpers
messages/                English and Arabic translations
public/                  Public static assets
scripts/                 Reproducible asset generation scripts
deploy/website/          Server runtime files copied during deployment
docs/                    Contributor and operational documentation

Local Development

Requirements:

• Node.js 24+
• pnpm 11+

Install dependencies:

corepack enable pnpm
pnpm install

Start the development server:

pnpm dev

Open:

http://localhost:3000

Application source code lives under src/.

Environment

Copy .env.example when local environment values are needed:

cp .env.example .env.local

Variable	Required	Purpose
NEXT_PUBLIC_SITE_URL	Local only	Canonical site URL for local testing
NEXT_PUBLIC_DATASETS_API_URL	Local only	API origin for future website/API integration
NEXT_PUBLIC_GOOGLE_TAG_MANAGER_ID	Production	Real Google tag ID, such as GT-WPDWW3NR, or Tag Manager container ID, such as GTM-ABC1234

The Google tag ID is public by design. Production passes it at Docker build time and writes it into the runtime environment. Use the real ID from Google, not a placeholder such as GTM-... or GTM-XXXXXXX.

Internationalization

The site supports English and Arabic.

• English is the default locale and renders at /.
• Arabic renders at /ar.
• Locale prefixes use next-intl as-needed routing.
• The HTML dir attribute and Base UI DirectionProvider are both driven from src/i18n/routing.ts.

Translations live in messages/en.json and messages/ar.json.

Social Previews

Open Graph and Twitter preview metadata explicitly reference the root social images:

• /opengraph-image.png
• /twitter-image.png

Keep these images aligned with the current OpenSyria logo and visual theme so crawlers do not fall back to contributor avatars or other page images. When replacing the image content, bump the brand asset version in src/lib/site.ts so social crawlers request the refreshed URLs.

Regenerate social preview assets after brand changes:

pnpm images:brand

Analytics

Google Tag Manager is integrated through @next/third-parties/google.

Tracked CTA events use:

event=cta_click
cta_id=docs
cta_id=github_stars

The implementation keeps the landing page server-rendered and uses small client boundaries for tracked links and interactive FAQ controls.

Checks

Run all CI checks:

pnpm verify

Focused commands:

pnpm check
pnpm typecheck
pnpm build

Apply Biome formatting and safe fixes:

pnpm check:write

Deployment

Production deployment is handled by .github/workflows/deploy-production.yml.

The workflow:

1. Runs pnpm verify.
2. Builds a Linux ARM64 Docker image from the runtime target.
3. Pushes SHA and main tags to GitHub Container Registry.
4. Joins the Tailscale network.
5. Copies deploy/website to /srv/opensyria/website.
6. Writes the server .env.
7. Runs the blue/green deployment script.

Runtime image tags:

ghcr.io/open-syria/website:sha-<short-sha>
ghcr.io/open-syria/website:main

See docs/deployment.md and deploy/website/README.md.

Production Routing

Cloudflare should route:

opensyria.org     -> http://website-proxy:80
api.opensyria.org -> http://proxy:80

Keep opensyria.org as the canonical host. Redirect http://opensyria.org, http://www.opensyria.org, and https://www.opensyria.org to https://opensyria.org in one Cloudflare redirect rule to avoid redirect chains.

Repository Documents

• Contributing
• Code of Conduct
• Security Policy
• Support
• Changelog
• Pull Request Workflow
• Supply Chain Security
• Deployment
• Releases

Contribution Model

The website is public for transparency, auditability, and reuse, but broad implementation work is maintainer-led.

Good public contributions here include documentation corrections, broken links, accessibility fixes, reproducible website bugs, deployment/tooling fixes, and maintainer-requested changes.

Dataset corrections belong in the relevant dataset repository.

License

Website code is licensed under MIT. See LICENSE.