horizon-rl/strands-env

Standardizing environment infrastructure with Strands Agents — step, observe, reward.

46
/ 100
Emerging

Treats each environment step as a full multi-turn agent loop with tool use, enabling token-level observations for on-policy RL training. Provides a subclassable `Environment` base with `@tool` decorators for defining custom actions, integrated reward functions, and a CLI-driven evaluator for benchmarking with checkpointing and resume. Directly integrates with Strands Agents SDK and strands-sglang for language model-based RL workflows.

No Package No Dependents
Maintenance 13 / 25
Adoption 8 / 25
Maturity 11 / 25
Community 14 / 25

How are scores calculated?

Stars

43

Forks

7

Language

Python

License

Apache-2.0

Last pushed

Mar 13, 2026

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/agents/horizon-rl/strands-env"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.