mdp-designer

What problem does it solve?

Designing and editing MDP terms (observations, rewards, terminations, goals/commands, and randomization) and wiring them into configs and logging to improve an RL environment's learning signal.

Core Features & Use Cases

• Identifies the term boundaries in an RL env and maps changes to the corresponding module (observations, rewards, terminations, goals, randomization).
• Provides a minimal, config-driven approach that preserves backwards compatibility while enabling targeted tweaks and clear logging.
• Supports a sanity-check plan (smoke run, unit checks) to validate that changes produce stable learning signals.

Quick Start

Run a smoke test by adjusting the MDP term boundary (e.g., observations or rewards) in your env and executing a short rollout to verify behavior.