OpenAgentSkill Registry Manifest
Skill: rl-reward
Slug: agentscope-ai-rl-reward
Category: design-creative
Description: Build RL reward signals using the OpenJudge framework. Covers choosing between pointwise and pairwise reward strategies based on RL algorithm, task type, and cost; aggregating multi-dimensional pointwise scores into a scalar reward; pairwise tournament reward for GRPO on subjective tasks (net win rate across group rollouts); generating preference pairs for DPO/RLAIF; and normalizing scores for training stability. Use when building reward models, scoring rollouts for GRPO/REINFORCE, generating preference data for DPO, or doing Best-of-N selection.

Agent fit:
- Decision: 81/100 Strong shortlist
- Primary fit: RAG and knowledge
- Role: Companion skill

Supply profile:
- Track: Design and creative production
- Scenario: Design and creative
- Applicable agents: Claude Code, CLI, Codex, Cursor
- Maintenance: 1mo since push
- Risk: Needs review

Trust:
- Trust score: 78/100 Strong shortlist
- Audit: 79/100 Needs review

Attribution:
- Status: Registry indexed
- Source: github fast track
- Creator: agentscope-ai
- Claim URL: https://www.openagentskill.com/skills/agentscope-ai-rl-reward#claim-this-skill

Install:
npx skills add agentscope-ai/OpenJudge --skill rl-reward

URLs:
- Web: https://www.openagentskill.com/skills/agentscope-ai-rl-reward
- API: https://www.openagentskill.com/api/agent/skills/agentscope-ai-rl-reward
- Install API: https://www.openagentskill.com/api/skills/agentscope-ai-rl-reward/install
- Repository: https://github.com/agentscope-ai/OpenJudge/tree/main/skills/rl-reward