apurvaga/textcraft-synth
TextCraft-Synth TextCraft-Synth is a text crafting game for training and evaluating LLM agents on long-horizon planning: craft the target items from a starting inventory of base materials, through recipes up to 12 levels deep. This repository holds it as an RL environment in the [Harbor][harbor] task format: 3,154 tasks, in each of which the agent plays through a textcraft command in a container, and a verifier replays the agent's moves to grade them. The tasks, rules… See the full description on the dataset page: https://huggingface.co/datasets/apurvaga/textcraft-synth.
| Author / organization | apurvaga |
|---|---|
| License (as declared) | mit |
| Downloads | 0 (as reported by Hugging Face) |
| Likes | 0 |
| Created | 16 October 2026 |
| Last updated at source | 16 October 2026 |
Connected in the knowledge graph
- Recursive Agent OptimizationPaper · cites
Recent changes
- 16 Oct 2026 License: none → mit
- 16 Oct 2026 Description: none → TextCraft-Synth TextCraft-Synth is a text crafting game for training and evaluating LLM agents on long-horizon planning: craft the target items from a starting inventory of base materials, through recipes up to 12 levels deep. This reposito
- 16 Oct 2026 Topics added rl-environment, harbor, textcraft, crafting, planning
Changes detected by Silytics between refreshes of the source.
Related papers
More from apurvaga
Silytics shows public metadata from Hugging Face and links to the original. Information may be incomplete or out of date. Last refreshed 16 October 2026. Licenses and terms of the original apply to the work itself. Not financial, legal or professional advice.