Skip to content
Back to skills

Training Integrations

ASecurity

"Adapt PettingZoo training tutorials for CleanRL, Tianshou, SB3,

  • 247 stars
  • 0 votes
  • 0 copies
  • 2 views
  • Added September 8, 2026
testingtestingapi

Works with

  • api

Security analysis

A100/100

Pro scans all 5 files and shows the line behind each finding

Scanned September 8, 2026

npx -y skills add VectorSpaceLab/AREX-Skill --skill training-integrations --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Training Integrations?

Add the live security badge to your README. It updates with every re-scan.

Security grade badge for Training Integrations
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/vectorspacelab-training-integrations/badge)](https://www.skillsdirectory.com/skills/vectorspacelab-training-integrations)

More formats (shields.io, HTML) on the badges page. Keep it an A: scan every change in CI with Pro.

Download with Pro
SKILL.md
---
name: training-integrations
description: "Adapt PettingZoo training tutorials for CleanRL, Tianshou, SB3,
  Ray/RLlib, AgileRL, and LangChain while keeping heavy execution gated."
disable-model-invocation: true
metadata:
  disco-role: operating
license: NOASSERTION
---

# PettingZoo Training Integrations

Use this sub-skill when a user asks to connect PettingZoo environments to downstream training or agent frameworks such as CleanRL, Tianshou, Stable-Baselines3, Ray/RLlib, AgileRL, or LangChain. Keep actual training, framework installation, GPU use, display rendering, credentialed services, and network activity opt-in.

## Read First

- Read [framework-recipes.md](references/framework-recipes.md) to choose the right framework recipe, environment family, dependency group, and safe adaptation path.
- Read [vectorization-and-action-masks.md](references/vectorization-and-action-masks.md) when converting between AEC and Parallel APIs, preserving action masks, or using SuperSuit/vector wrappers.
- Read [troubleshooting.md](references/troubleshooting.md) when imports, version constraints, masks, vectorization, rendering, or long-running training fail.
- Run [inspect_integration_requirements.py](scripts/inspect_integration_requirements.py) to print the known tutorial requirement groups and optionally check whether framework modules are importable locally without installing anything.

## Routing Boundaries

- For basic PettingZoo `AECEnv` and `ParallelEnv` loops, route to `../use-environments/SKILL.md`.
- For optional family extras, ROMs, rendering dependencies, and environment selection, route to `../environment-families/SKILL.md`.
- For conversion wrappers, `AgentSelector`, utility wrappers, and wrapper ordering, route to `../wrappers-and-utilities/SKILL.md`.
- For compliance tests or safe native-test selection, route to `../testing-and-validation/SKILL.md`.

## Default Safety Policy

Training tutorials are reference recipes, not default runtime scripts. Before running any training framework command, confirm the user wants the required environment mutation, dependency installation, compute budget, checkpoint writes, display rendering, network access, or credentials.

Files in this skill

  • SKILL.md2.2 KB
  • references/framework-recipes.md12.8 KB
  • references/troubleshooting.md6.8 KB
  • references/vectorization-and-action-masks.md7.1 KB
  • scripts/inspect_integration_requirements.py9.8 KB

Attribution

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.

Comments

Loading comments…

Newsletter

Get the weekly list of safe new skills, and the ones we pulled.