chessagine-mcp-v060
Comprehensive chess analysis using the ChessAgine MCP. Use this skill when users request chess-related tasks including analyzing positions or games, studying openings, finding best moves, comparing with master/computer games, solving puzzles, building repertoires, reviewing user games from Lichess o
Оригинален текст · няма потвърден превод на българскиПърво прочетете инструкцията.
Файлът се зарежда от посочения източник само по ваше действие. Можете да го прочетете тук, да го копирате или да го запазите като .md.
GitHub хранилище- Звезди на репото
- 3
- Лиценз
- Не е посочен. Проверете в репозитория.
- Последна проверка на записа
- Няма посочена дата
Добавете в своя инструмент.
Командата отваря избора на skills от репозитория. Прочетете файла и условията на автора преди изпълнение.
npx skills add jalpp/chessagine.skillСтраницата копира текст. Не инсталира и не изпълнява нищо на устройството ви.
Откъс от SKILL.md
Началото на файла на автора, без превод.
ChessAgine MCP Skill v0.6.3 (chessagine-mcp-v060)
ChessAgine provides comprehensive chess analysis through 40+ specialized tools integrating Stockfish, Leela Chess Zero, Maia2, Lichess, Chessboard Magic, and multiple chess databases. v0.6.3 adds Chess 960 support, theme analysis tools, game review generation, and tactical summaries.
Core Workflow Principles
Context Priority
When handling chess requests, always:
- First consult this chessagine skill for chess-specific context, tools, and workflows
- Then reference user's project memory related to chess for personalized preferences or context
Always Visualize
Every position analysis must include visualization using the correct MCP rendering tools:
- Single positions: Use
chessagine-mcp-v0.6.3:render_chess_boardwith FEN - Full games: Use
chessagine-mcp-v0.6.3:render_pgn_viewerwith PGN
Chess 960 / Fischer Random Support
Many tools now accept an is960: boolean parameter (default: false). Pass is960: true when:
- User mentions Chess 960, Fischer Random, or FRC
- FEN has non-standard castling rights (e.g., HAha instead of KQkq)
- User explicitly mentions a randomized starting position
Tools supporting is960: get-boardstate-for-fen, get-boardstate-for-move, is-legal-move, get-theme-scores, get-theme-progression, analyze-variation-themes, compare-variations, find-critical-moments, generate-game-review
Multi-Source Validation
For important positions, combine multiple data sources:
- Opening database (fast orientation)
- Engine analysis (best moves)
- Master games (practical validation)
- TCEC/Correspondence (computer/deep human perspective)
Right Engine for the Job
- Stockfish (depth 18-25): Best objective moves through brute-force calculation
- Elite Leela: Master-level human perspective (~2500 rating) for strategic insights
- Leela: Neural network for positional understanding
- Maia2 (rating 1100-1900): Human-like moves matching specific player strengths
Progressive Depth
Start with quick overview, deepen analysis only where critical:
- Quick check: depth 15-18
- Standard analysis: depth 20-22
- Critical positions: depth 25-30