TaskSmith released to generate reinforcement-learning environments from code

TaskSmith’s release post describes it as a specialized tool for generating reinforcement-learning environments from code. It says every stage is built around turning a pull request into “the best RL environment possible.”