games.chess.engines¶

Chess game engines using AugLLMConfig.

This module provides LLM engine configurations for chess game agents, including:

  • Player engines for white and black

  • Analyzer engines for position evaluation

  • Prompt templates with chess-specific instructions

  • Structured output models for moves and analysis

The engines use different LLM configurations optimized for their specific roles, with prompt templates designed to generate high-quality chess moves and analysis.

Functions¶

build_chess_aug_llms()

Build AugLLMConfig dictionary for chess game engines.

create_black_analyzer_engine()

Create black analyzer engine configuration.

create_black_player_engine()

Create black player engine configuration.

create_white_analyzer_engine()

Create white analyzer engine configuration.

create_white_player_engine()

Create white player engine configuration.

Module Contents¶

games.chess.engines.build_chess_aug_llms()¶

Build AugLLMConfig dictionary for chess game engines.

Creates a complete set of engine configurations for a chess agent, including players and analyzers for both white and black.

Returns:

Dictionary of engine configurations with keys:

  • ”white_player”: Engine for white’s moves

  • ”black_player”: Engine for black’s moves

  • ”white_analyzer”: Engine for analyzing positions from white’s perspective

  • ”black_analyzer”: Engine for analyzing positions from black’s perspective

Return type:

dict[str, AugLLMConfig]

Examples

>>> engines = build_chess_aug_llms()
>>> len(engines)
4
>>> sorted(list(engines.keys()))
['black_analyzer', 'black_player', 'white_analyzer', 'white_player']
games.chess.engines.create_black_analyzer_engine()¶

Create black analyzer engine configuration.

Configures an LLM engine for analyzing chess positions from black’s perspective, providing structured analysis with: - Position evaluation score - Attacking opportunities - Defensive needs - Strategic plans

Returns:

Configuration for the black analyzer engine

Return type:

AugLLMConfig

Examples

>>> engine = create_black_analyzer_engine()
>>> engine.name
'black_analyzer'
>>> engine.structured_output_model
<class 'haive.games.chess.models.SegmentedAnalysis'>
games.chess.engines.create_black_player_engine()¶

Create black player engine configuration.

Configures an LLM engine for generating black’s chess moves with: - Specific instructions for UCI move format - Examples of valid moves - Structured output using ChessPlayerDecision model - Appropriate temperature for strategic play

Returns:

Configuration for the black player engine

Return type:

AugLLMConfig

Examples

>>> engine = create_black_player_engine()
>>> engine.name
'black_player'
>>> engine.structured_output_model
<class 'haive.games.chess.models.ChessPlayerDecision'>
games.chess.engines.create_white_analyzer_engine()¶

Create white analyzer engine configuration.

Configures an LLM engine for analyzing chess positions from white’s perspective, providing structured analysis with: - Position evaluation score - Attacking opportunities - Defensive needs - Strategic plans

Returns:

Configuration for the white analyzer engine

Return type:

AugLLMConfig

Examples

>>> engine = create_white_analyzer_engine()
>>> engine.name
'white_analyzer'
>>> engine.structured_output_model
<class 'haive.games.chess.models.SegmentedAnalysis'>
games.chess.engines.create_white_player_engine()¶

Create white player engine configuration.

Configures an LLM engine for generating white’s chess moves with: - Specific instructions for UCI move format - Examples of valid moves - Structured output using ChessPlayerDecision model - Appropriate temperature for strategic play

Returns:

Configuration for the white player engine

Return type:

AugLLMConfig

Examples

>>> engine = create_white_player_engine()
>>> engine.name
'white_player'
>>> engine.structured_output_model
<class 'haive.games.chess.models.ChessPlayerDecision'>