0751d988c3
- Restore all 12,379 openings from commit 8ed402b
- Rebuild move index with 12,377 sequences
- Fix merge issue that overwrote with test fixtures
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com>
Opening Database
This directory contains the ECO (Encyclopedia of Chess Openings) database files.
Required Files
The following files are required for opening training:
ecoA.json- ECO codes A00-A99ecoB.json- ECO codes B00-B99ecoC.json- ECO codes C00-C99ecoD.json- ECO codes D00-D99ecoE.json- ECO codes E00-E99moveIndex.json- Generated move sequence index
File Format
Each ECO file (ecoA-E.json) should be a JSON object mapping FEN positions to opening metadata:
{
"fen_position": {
"eco": "A00",
"name": "Opening Name",
"moves": "e4 e5 Nf3 Nc6",
"wikipediaSlug": "opening-name" // optional
}
}
Setup
Option 1: Docker (Automatic)
When running via Docker, these files should be provided as a volume mount:
docker run -v ./openings:/app/public/openings ghcr.io/stefan-kp/chess-tutor
Option 2: Local Development
- Obtain ECO database files (ecoA-E.json)
- Place them in this directory
- Generate the move index:
npm run build:opening-index
This will create moveIndex.json from the ECO files.
Option 3: Generate from PGN
If you have a PGN database, you can extract ECO codes using chess tools like:
pgn-extract- Custom scripts
Notes
- These files are not included in git (too large, ~4MB total)
- Users must provide their own opening database
- Wikipedia integration is optional (see
public/wikipedia/README.md) - The move index is automatically generated during Docker startup