I built my family a private Chess.com — one Docker container, Stockfish plus your own LLM

Description and use

Patzer is a local, family‑friendly chess platform running in a single Docker container, using Stockfish for engine analysis and a local LLM for coaching, eliminating external accounts and ads.

Problem

Parents want a child‑friendly chess platform without external accounts, cloud storage, or ads.

Outcome

Patzer delivers a fully local, family‑friendly chess environment with engine‑based analysis and LLM‑powered coaching, eliminating external data exposure and ads.

Workflow

• Pull public Chess.com or Lichess games or paste a PGN
• Stockfish classifies each move and provides accuracy, Elo, eval graph, and plain‑language threat analysis
• Play against Stockfish at seven difficulty tiers with kid‑mode blunder warnings
• Play real‑time PvP over WebSocket with draw offers, takebacks, and rematch
• Point AI coach to a local Ollama or vLLM host for move explanations and mistake analysis
• Configure multi‑user profiles, languages, kid mode, and custom piece names via admin console

Evidence

My kids play chess on Chess.com. It is a great product, but the account, the cloud and the ads are not something I want to hand to a nine-year-old.

Source

One container: docker run -d -p 8800:8800 -v patzer-data:/app/data \
--name patzer ghcr.io/SikamikanikoBG/patzer:latest

Source

The server works out the facts — what a move allows, hangs or wins, the opponent's best answer, your recurring mistakes — and the model only turns that into coaching.

Source