Safe Rl Python Packages

ghostloop

The agent loop, embodied. Tool-using runtime + fail-closed safety pipeline + sim-first execution + statistically-rigorous bench harness + post-hoc analysis (counterfactual replay, causal attribution, LLM judge, property mining) for embodied AI. Sister project to GhostLM.

omnisafe

A comprehensive and reliable benchmark for safe reinforcement learning.

parallelcbf

Composable safety-filter and auditability layer for tensor-parallel reinforcement learning. CBF safety filters, watchdog registries, pre-registration artifacts, failure forensics.

dsrl

Datasets for Offline Safe Reinforcement Learning

fast-safe-rl

A Fast and Modularized Implementation of Safe Reinforcement Learning Algorithms in PyTorch

osrl-lib

🤖 Elegant implementations of offline safe RL algorithms in PyTorch