skills.rest
junhyekh
junhyekhCOMMUNITY
Explore all skills by this author →
GitHub

rl-reward-curriculum-design

Route legacy RL reward and curriculum requests to RLXP intervention skills.

1|Updated May 14, 2026
Search Tags:#curriculum-design#experiment-planning#domain-randomization#rlxp#reward-tuning#candidate-validation#compatibility-wrapper
One-click install
npx skills add https://github.com/junhyekh/rlxp --skill rl-reward-curriculum-design
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: rl-reward-curriculum-design
Source: https://github.com/junhyekh/rlxp/tree/main/plugins/rl-experiment-assistant/skills/rl-reward-curriculum-design
Command: npx skills add https://github.com/junhyekh/rlxp --skill rl-reward-curriculum-design

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This legacy wrapper keeps older reward, curriculum, and domain-randomization prompts working by directing them to the narrower RLXP skills that handle each task safely and consistently.

Core Features & Use Cases

  • Dispatches bounded reward tuning to the reward-tuning skill.
  • Sends reward code changes to reward-engineering only when evidence and contract approval justify it.
  • Covers curriculum design, DR probing, reward review, and candidate validation without expanding scope.
  • Use Case: A researcher asks for help adjusting reward terms for a new behavior, and the skill routes the request to the right intervention path while preserving guardrails.

Quick Start

Ask for legacy RL reward or curriculum help, and let the skill route your request to the correct RLXP intervention path.

Frequently Asked Questions about rl-reward-curriculum-design

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

FAQPage Schema
How do I safely route legacy reward tuning requests for reinforcement learning experiments?▼

Legacy reward tuning requests are routed to dedicated reward-tuning intervention paths, ensuring bounded experiment planning while preserving held-out evaluation semantics and enforcing no-training boundaries. This dispatches older prompts safely without expanding scope.

What is domain randomization probing and when do I need it for RL experiment planning?▼

Domain randomization probing evaluates environment variability constraints within bounded experiment planning. It is needed when adapting reinforcement learning policies to simulated conditions, routing legacy DR requests safely through validated intervention paths without violating active contracts.

Can I use this to update reward code changes for new behaviors in my RL environment?▼

Reward code changes are dispatched to reward-engineering paths only when evidence and contract approval justify them. This ensures modifications for new behaviors adhere to active contracts and launch gates while maintaining no-training boundaries.

How do I design a curriculum for reinforcement learning without breaking held-out evaluation semantics?▼

Curriculum design requests are routed to specific intervention paths that enforce launch gates and no-training boundaries. This dispatches legacy curriculum prompts safely, ensuring experiment planning remains bounded and held-out evaluation semantics are preserved.

What are the limitations of using a compatibility wrapper for legacy reinforcement learning prompts?▼

The compatibility wrapper limits scope by strictly routing legacy prompts to existing intervention paths without expanding functionality. It requires adherence to active contracts and launch gates, meaning it cannot bypass no-training boundaries or held-out evaluation constraints.

Related Skills

Explore More Software Engineering Skills→
itallstartedwithaideaitallstartedwithaidea
31

machine-learning

Build reproducible PyTorch and scikit-learn pipelines with stratified cross-validation and Optuna tuning.

Community
Advanced
junhyekhjunhyekh
1

rlxp-plan-baseline

Create a study-scoped reinforcement-learning baseline plan with training commands and metric paths.

Community
Advanced
junhyekhjunhyekh
1

rlxp-reward-reflection

Generate a reward reflection JSON with component diagnosis and next action.

Community
Advanced
pdparchitectpdparchitect
13

hunt-auth-bypass

Identify and validate authentication bypass vulnerabilities across SSO, SAML, OAuth, JWT, and legacy endpoints.

Community
Advanced