Anarch Chess

Client-side implementation of Anarch Chess. The board rules, AI search, evaluation, post-game analysis, and AI worker run in the browser. Match archive requests use /api when a backend is available and fall back to localStorage when it is not.

Local development

npm ci
npm run dev

The frontend-only command is useful when you do not need match persistence. For the complete local application, including the reconstructed TypeScript SQLite API:

npm run build
npm run server

Then open http://127.0.0.1:8787/. The server serves the production frontend, the /api/matches and /api/stats endpoints, and creates data/anarch-chess.sqlite automatically. Node 24 includes the SQLite driver, so SQLite does not need to be installed separately.

During UI development, run npm run dev and npm run server in separate terminals. Vite proxies /api to port 8787.

To run the compiler checks separately:

npm run typecheck

Netlify

Create a Netlify site from this repository with the base directory set to anarch-chess. The included netlify.toml builds the Vite application and rewrites SPA routes to index.html.

Equivalent settings:

• Build command: npm run build
• Publish directory: dist
• Base directory: anarch-chess

Render

Create a Render Web Service using the included render.yaml. It serves the built frontend and the SQLite API from one Node process:

• Root directory: anarch-chess
• Build command: npm ci && npm run build
• Start command: npm run server
• Node version: 24
• DATA_DIR: ./data

The Node server rewrites unknown frontend GET routes to index.html, so direct navigation to /rules and /about works. For production persistence, attach a Render persistent disk at /opt/render/project/src/anarch-chess/data; otherwise the database is reset when the service is redeployed.