# microsoft/TextWorld

​TextWorld is a sandbox learning environment for the training and evaluation of reinforcement learning (RL) agents on text-based games.

Repository: https://github.com/microsoft/TextWorld
Canonical: https://ross.abutalabs.com/products/textworld
Language: Jupyter Notebook
License: NOASSERTION
License Family: other
Topics: text-based-game, reinforcement-learning, text-based-adventure
Last push: 2026-01-30T19:15:32+00:00

## Health v2 (maintenance only)
Score: 62/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 65, release rhythm 36, longevity 100
- inputs: {"age_days": 2990, "days_push": 215, "days_rel": 215, "gap_med": null, "n_releases_24m": 1}
- flags: no_license
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1441, forks 205 (observed 2026-08-28T04:04:44.321094+00:00)

## What it is
TextWorld is a Python library from Microsoft that generates text-based adventure games and provides a sandbox learning environment for training and evaluating reinforcement learning agents. It supports game state visualization and integrates with standard RL workflows.

## Use cases
- train reinforcement learning agents on text-based games
- generate text adventure games programmatically
- benchmark RL agents on text environments
- research NLP and RL in interactive fiction settings
- visualize game states for text-based environments

## When to choose
- you need a reproducible sandbox of text games for RL research
- you want procedurally generated text adventures with controllable difficulty
- you work in Python on Linux or macOS

## When to avoid
- you need native Windows support without Docker
- you want graphical (non-text) game environments
- you need a production game engine rather than a research tool

## Facets
- artifact type: library
- maturity: active
- function: reinforcement-learning, game-engine, simulation, nlp
- domain: reinforcement-learning, artificial-intelligence
- platform: python
- tags: text-based-games, rl-environment, gym-environment, game-generator, sandbox, game-development, natural-language-processing, linux, macos, docker

## Member repositories
- microsoft/TextWorld (main) score 62

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:04:44.321094+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T04:36:28.774233+00:00, confidence not recorded.
  - readme: https://github.com/microsoft/TextWorld (fetched 2026-08-28T04:04:44.321094+00:00, sha 59e3ca5e5763)
  - registry_pypi: https://pypi.org/pypi/textworld/json (fetched 2026-08-29T11:46:40.422325+00:00, sha c17af3a7e457)
- Data as of 2026-08-30T08:39:29.467469+00:00.