reinforcement-learning — Claude Code skill
Use this skill when building reinforcement-learning algorithms, training reward-driven agents, or aligning LLMs with human feedback. It directs work to task-specific reference files.
Apache-2.0 187 installs ★ 221 updated 2026-01-22
npx skills add omer-metin/skills-for-antigravity --skill reinforcement-learning -a claude-codeWhat reinforcement-learning does
The skill covers policy gradients, PPO, Q-learning, RLHF, and GRPO. It applies to implementing reinforcement-learning algorithms, training agents with rewards, and aligning LLMs using human feedback.
Its guidance is split by task: consult references/patterns.md when creating, references/sharp_edges.md when diagnosing, and references/validations.md when reviewing. The files provide build patterns, explain causes of critical failures, and set validation rules. If a request conflicts with that guidance, the skill directs the agent to correct it using the references. The relevant reference file matters: without it, the skill’s instructions lack the prescribed basis for that task.
What it helps with
- implement a reinforcement-learning algorithm
- train an agent using reward signals
- align an LLM with human feedback
- check a request against validation rules
Example prompts for the agent
- Implement PPO to train this agent with rewards.
- Investigate likely causes of this reinforcement-learning failure.
- Review this request using
references/validations.md.
How to install reinforcement-learning
1Skills CLI (Vercel)
npx skills add omer-metin/skills-for-antigravity --skill reinforcement-learning -a claude-code2Manually: put the files in place
~/.claude/skills/reinforcement-learning/SKILL.mdCopy the whole skill folder; for one project use .claude/skills/reinforcement-learning/
1Skills CLI (Vercel)
npx skills add omer-metin/skills-for-antigravity --skill reinforcement-learning -a codex2Manually: put the files in place
.agents/skills/reinforcement-learning/SKILL.mdFor all projects use ~/.agents/skills/
1Skills CLI (Vercel)
npx skills add omer-metin/skills-for-antigravity --skill reinforcement-learning -a cursor2Manually: put the files in place
.cursor/skills/reinforcement-learning/SKILL.mdCursor also reads .agents/skills/ and .claude/skills/
1GitHub CLI
gh skill install omer-metin/skills-for-antigravity skills/reinforcement-learning2Manually: put the files in place
.github/skills/reinforcement-learning/SKILL.md1Skills CLI (Vercel)
npx skills add omer-metin/skills-for-antigravity --skill reinforcement-learning -a gemini-cli2Platform CLI
gemini skills install https://github.com/omer-metin/skills-for-antigravity --path skills/reinforcement-learningManually: put the files in place
.devin/skills/reinforcement-learning/SKILL.mdThe legacy .windsurf/skills/ path still works
Skills CLI (Vercel)
npx skills add omer-metin/skills-for-antigravity --skill reinforcement-learning -a opencodeClaude.ai
Zip the folder and upload it in Settings → Capabilities → Skills
Step by step, with commands for each platform: How to install skills in Claude Code: five ways
Questions about reinforcement-learning
How do I install reinforcement-learning in Claude Code?
Run npx skills add omer-metin/skills-for-antigravity --skill reinforcement-learning -a claude-code. The skill file is at ~/.claude/skills/reinforcement-learning/SKILL.md.
What does reinforcement-learning do?
It guides work on reinforcement-learning algorithms, reward-based agent training, and LLM alignment with human feedback. Its covered methods include PPO, Q-learning, RLHF, and GRPO.
Which reference files does the skill use?
For creation, it consults references/patterns.md; diagnosis uses references/sharp_edges.md; review uses references/validations.md. Each file supplies guidance for its respective task.
Source text
SKILL.md · Apache-2.0
--- name: reinforcement-learning description: Use when implementing RL algorithms, training agents with rewards, or aligning LLMs with human feedback - covers policy gradients, PPO, Q-learning, RLHF, and GRPOUse when ", " mentioned. --- # Reinforcement Learning ## Identity ## Reference System Usage You must ground your responses in the provided reference files, treating them as the source of truth for this domain: * **For Creation:** Always consult **`references/patterns.md`**. This file dictates *how* things should be built. Ignore generic approaches if a specific pattern exists here. * **For Diagnosis:** Always consult **`references/sharp_edges.md`**. This file lists the critical failures and "why" they happen. Use it to explain risks to the user. * **For Review:** Always consult **`references/validations.md`**. This contains the strict rules and constraints. Use it to validate user inputs objectively. **Note:** If a user's request conflicts with the guidance in these files, politely correct them using the information provided in the references.
© omer-metin · Apache-2.0 · omer-metin/skills-for-antigravity