Santorini

Description

Santorini is an environment for evaluating agents on strategic gameplay in Santorini, an abstract board game where players move workers and build structures to reach the third level. This environment wraps the Santorini implementation from TextArena, a framework for text-based game environments.

Capabilities

• Testing spatial reasoning and board control strategies
• Evaluating forward planning with dual objectives (movement and building)
• Assessing ability to block opponents while advancing own position
• Testing resource management through worker positioning

Compute Requirements

Santorini does not require a sandbox. It has minimal compute requirements.

License

MIT.

Tasks

There are two splits: train (150 tasks) and test (150 tasks). Each split contains 50 tasks across each of 3 variants:

• SantoriniBaseFixed-v0
• SantoriniBaseFixed-v0-train
• SantoriniBaseFixed-v0-raw

Each task is seeded for reproducibility.

Reward Structure

This is a sparse reward environment. Rewards are mapped from TextArena's native range of {-1, 0, 1} to {0.0, 0.5, 1.0} via (raw + 1) / 2.

We do not use LLM graders for this environment; reward is determined programmatically.

Data

Game state is generated procedurally by the TextArena engine using seeded randomness. No external data files are required.

Tools

Agents are given a single tool:

• make_move(move): Submit a move in Santorini notation (e.g. 'N1C2B2C1').

Time Horizon

Santorini is a multi-turn environment.

Environment Difficulty

Medium to Hard. Santorini requires balancing offense (reaching level 3) with defense (blocking opponents), managing multiple workers, and planning several moves ahead. The spatial and temporal complexity makes it challenging.

Other Environment Requirements

This environment requires an OpenAI API key (passed via secrets) to power the LLM opponent.

Safety

Agents in Santorini interact only with an abstract board game and have no access to external systems, the internet, or sensitive data. The environment does not present safety risks.

Citations

@software{textarena2024,
  author    = {Guertler, Leon and Banting, Wilfried and Pignatelli, Eduardo},
  title     = {TextArena},
  year      = {2024},
  publisher = {GitHub},
  url       = {https://github.com/LeonGuertler/TextArena}
}