← All skills

grpo-rlvr-training

95 / 100 · Low risk

Fine to try. Some of what it can do is expected: a coding skill usually runs build and test commands.

By wshobson/agents (GitHub) · coding · 42,412 stars · scanned at commit 9c61ab2

Train reasoning and verifiable-task behavior with GRPO and reinforcement learning from verifiable rewards (RLVR). Use when task success is algorithmically checkable (math, code, tool calls, structured output), when designing GRPO reward functions, or when a GRPO run diverges or reward-hacks.

Works with any agent that supports SKILL.md skills.

What this skill can do

Install

Paste this into your terminal. It downloads exactly the version we scanned, not whatever is newest.

if [ -e ~/.agents/skills/'grpo-rlvr-training' ] || [ -L ~/.agents/skills/'grpo-rlvr-training' ]; then echo 'A skill folder named grpo-rlvr-training already exists — not replacing it' >&2; false; else d=$(mktemp -d) && git clone --quiet --filter=blob:none --no-checkout 'https://github.com/wshobson/agents' "$d" && git -C "$d" checkout --quiet '9c61ab243def98f72da12abd41cfcf5bcd91af9d' -- 'plugins/llm-finetuning/skills/grpo-rlvr-training' && mkdir -p ~/.agents/skills && cp -R "$d"/'plugins/llm-finetuning/skills/grpo-rlvr-training' ~/.agents/skills/'grpo-rlvr-training' && rm -rf "$d"; fi

Codex reads ~/.agents/skills directly. For Claude Code, Cursor, Copilot, Gemini CLI, Antigravity or OpenCode, also link it. For Claude Code: ln -s ~/.agents/skills/'grpo-rlvr-training' ~/.claude/skills/'grpo-rlvr-training' (other agents: ~/.cursor/skills, ~/.copilot/skills, ~/.gemini/skills, ~/.gemini/config/skills, ~/.config/opencode/skills). Or install through the finder skill, which does both.

Badge for your README

The badge for wshobson/agents shows the score of its lowest-scoring skill.

[![SkillTrust](https://zalavijaysinh.github.io/skilltrust/badges/wshobson/agents.svg)](https://zalavijaysinh.github.io/skilltrust/repos/wshobson/agents/)

Score history