rl-environments

Design and validate custom Gym/Gymnasium environments with spaces, wrappers, and vectorization.

1|Updated Oct 29, 2025
One-click install
npx skills add https://github.com/tachyon-beep/hamlet --skill rl-environments
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: rl-environments
Source: https://github.com/tachyon-beep/hamlet/tree/main/.claude/skills/yzmir-deep-rl/skills/rl-environments
Command: npx skills add https://github.com/tachyon-beep/hamlet --skill rl-environments

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill helps developers design, validate, and debug custom Gym/Gymnasium environments, including spaces, wrappers, and vectorized configurations, to accelerate RL experiments and ensure API compatibility.

Core Features & Use Cases

  • Create and customize environments with proper action and observation spaces.
  • Apply wrappers for preprocessing, rewards, time limits, and parallelization.
  • Build and validate vectorized environments (DummySyncAsync) for scalable training.
  • Diagnose common environment issues (reset/step semantics, space mismatches, termination conditions) and migrate code between Gym and Gymnasium.
  • Use as a foundational blueprint for teaching or researching RL environment design patterns.

Quick Start

Use the rl-environments skill to scaffold a minimal CartPole-like environment, define discrete actions and Box observations, implement reset and step, then wrap with a TimeLimit and run a simple random-agent loop to verify shapes and termination.

Frequently Asked Questions about rl-environments

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I build a custom Gymnasium environment with correct action and observation spaces?

Building a custom Gymnasium environment requires defining action and observation spaces, implementing reset and step semantics, and applying wrappers like TimeLimit to validate shapes and termination conditions for training.

Why does my Gymnasium environment throw space mismatch errors during step or reset?

Space mismatches in Gymnasium environments happen when action or observation data violates defined space boundaries. Diagnose these issues by validating reset and step semantics, checking space definitions, and verifying termination conditions.

How do I migrate custom environments from Gym to Gymnasium?

Migrating environments from Gym to Gymnasium requires updating API compatibility, adjusting space definitions, and verifying reset and step semantics to ensure the custom environment meets Gymnasium API standards.

Can I use wrappers and vectorization for parallel training in Gymnasium environments?

You can apply wrappers for preprocessing, rewards, and time limits, and build vectorized environments like DummySyncAsync configurations to enable scalable parallel training in Gymnasium environments.

What is the best way to validate a custom Gym environment before running RL experiments?

The best way to validate a custom Gym environment is to scaffold a minimal loop with a random agent, apply a TimeLimit wrapper, and verify observation shapes, action validation, and termination conditions to ensure API compatibility.