Tak

Description

Tak is an environment for evaluating agents on the full version of Tak, a strategic board game of roads and stacks. This environment wraps the Tak implementation from TextArena, a framework for text-based game environments.

Capabilities

• Complex strategic board game playing
• Piece placement and stack manipulation
• Road-building and territory control
• Evaluation across medium and hard difficulty variants

Compute Requirements

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

License

MIT.

Tasks

There are two splits: train (450 tasks) and test (450 tasks). Each split contains 50 tasks across each of 9 variants:

• Tak-v0
• Tak-v0-hard
• Tak-v0-hard-raw
• Tak-v0-hard-train
• Tak-v0-medium
• Tak-v0-medium-raw
• Tak-v0-medium-train
• Tak-v0-raw
• Tak-v0-train

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 Tak notation.

Time Horizon

Tak is a multi-turn environment.

Environment Difficulty

Hard - requires deep strategic thinking and planning.

Other Environment Requirements

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

Safety

Agents in Tak interact only with a 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}
}