agentsclimarketplace
Sign in

The catalog

326,144 rows, by stacks then stars

  • Paper writing

    wanshuiyin/Auto-claude-code-research-in-sleep/skills/skills-codex-gemini-review/paper-writing Skill

    14,663★ repo

    ARIS ⚔️ (Auto-Research-In-Sleep) — Lightweight Markdown-only skills for autonomous ML research: cross-model review loops, idea discovery, and experiment automation. No framework, no lock-in — works with Claude Code, Codex, OpenClaw, or any LLM agent.

  • Research refine

    wanshuiyin/Auto-claude-code-research-in-sleep/skills/skills-codex-gemini-review/research-refine Skill

    14,663★ repo

    ARIS ⚔️ (Auto-Research-In-Sleep) — Lightweight Markdown-only skills for autonomous ML research: cross-model review loops, idea discovery, and experiment automation. No framework, no lock-in — works with Claude Code, Codex, OpenClaw, or any LLM agent.

  • Research review

    wanshuiyin/Auto-claude-code-research-in-sleep/skills/skills-codex-gemini-review/research-review Skill

    14,663★ repo

    ARIS ⚔️ (Auto-Research-In-Sleep) — Lightweight Markdown-only skills for autonomous ML research: cross-model review loops, idea discovery, and experiment automation. No framework, no lock-in — works with Claude Code, Codex, OpenClaw, or any LLM agent.

  • Ablation planner

    wanshuiyin/Auto-claude-code-research-in-sleep/skills/skills-codex/ablation-planner Skill

    14,663★ repo

    ARIS ⚔️ (Auto-Research-In-Sleep) — Lightweight Markdown-only skills for autonomous ML research: cross-model review loops, idea discovery, and experiment automation. No framework, no lock-in — works with Claude Code, Codex, OpenClaw, or any LLM agent.

  • Alphaxiv

    wanshuiyin/Auto-claude-code-research-in-sleep/skills/skills-codex/alphaxiv Skill

    14,663★ repo

    ARIS ⚔️ (Auto-Research-In-Sleep) — Lightweight Markdown-only skills for autonomous ML research: cross-model review loops, idea discovery, and experiment automation. No framework, no lock-in — works with Claude Code, Codex, OpenClaw, or any LLM agent.

  • Analyze results

    wanshuiyin/Auto-claude-code-research-in-sleep/skills/skills-codex/analyze-results Skill

    14,663★ repo

    ARIS ⚔️ (Auto-Research-In-Sleep) — Lightweight Markdown-only skills for autonomous ML research: cross-model review loops, idea discovery, and experiment automation. No framework, no lock-in — works with Claude Code, Codex, OpenClaw, or any LLM agent.

  • Arxiv

    wanshuiyin/Auto-claude-code-research-in-sleep/skills/skills-codex/arxiv Skill

    14,663★ repo

    ARIS ⚔️ (Auto-Research-In-Sleep) — Lightweight Markdown-only skills for autonomous ML research: cross-model review loops, idea discovery, and experiment automation. No framework, no lock-in — works with Claude Code, Codex, OpenClaw, or any LLM agent.

  • Auto paper improvement loop

    wanshuiyin/Auto-claude-code-research-in-sleep/skills/skills-codex/auto-paper-improvement-loop Skill

    14,663★ repo

    ARIS ⚔️ (Auto-Research-In-Sleep) — Lightweight Markdown-only skills for autonomous ML research: cross-model review loops, idea discovery, and experiment automation. No framework, no lock-in — works with Claude Code, Codex, OpenClaw, or any LLM agent.

  • Auto review loop llm

    wanshuiyin/Auto-claude-code-research-in-sleep/skills/skills-codex/auto-review-loop-llm Skill

    14,663★ repo

    ARIS ⚔️ (Auto-Research-In-Sleep) — Lightweight Markdown-only skills for autonomous ML research: cross-model review loops, idea discovery, and experiment automation. No framework, no lock-in — works with Claude Code, Codex, OpenClaw, or any LLM agent.

  • Auto review loop minimax

    wanshuiyin/Auto-claude-code-research-in-sleep/skills/skills-codex/auto-review-loop-minimax Skill

    14,663★ repo

    ARIS ⚔️ (Auto-Research-In-Sleep) — Lightweight Markdown-only skills for autonomous ML research: cross-model review loops, idea discovery, and experiment automation. No framework, no lock-in — works with Claude Code, Codex, OpenClaw, or any LLM agent.

  • Auto review loop

    wanshuiyin/Auto-claude-code-research-in-sleep/skills/skills-codex/auto-review-loop Skill

    14,663★ repo

    ARIS ⚔️ (Auto-Research-In-Sleep) — Lightweight Markdown-only skills for autonomous ML research: cross-model review loops, idea discovery, and experiment automation. No framework, no lock-in — works with Claude Code, Codex, OpenClaw, or any LLM agent.

  • Citation audit

    wanshuiyin/Auto-claude-code-research-in-sleep/skills/skills-codex/citation-audit Skill

    14,663★ repo

    ARIS ⚔️ (Auto-Research-In-Sleep) — Lightweight Markdown-only skills for autonomous ML research: cross-model review loops, idea discovery, and experiment automation. No framework, no lock-in — works with Claude Code, Codex, OpenClaw, or any LLM agent.

  • Claims drafting

    wanshuiyin/Auto-claude-code-research-in-sleep/skills/skills-codex/claims-drafting Skill

    14,663★ repo

    ARIS ⚔️ (Auto-Research-In-Sleep) — Lightweight Markdown-only skills for autonomous ML research: cross-model review loops, idea discovery, and experiment automation. No framework, no lock-in — works with Claude Code, Codex, OpenClaw, or any LLM agent.

  • Comm lit review

    wanshuiyin/Auto-claude-code-research-in-sleep/skills/skills-codex/comm-lit-review Skill

    14,663★ repo

    ARIS ⚔️ (Auto-Research-In-Sleep) — Lightweight Markdown-only skills for autonomous ML research: cross-model review loops, idea discovery, and experiment automation. No framework, no lock-in — works with Claude Code, Codex, OpenClaw, or any LLM agent.

  • Deepxiv

    wanshuiyin/Auto-claude-code-research-in-sleep/skills/skills-codex/deepxiv Skill

    14,663★ repo

    ARIS ⚔️ (Auto-Research-In-Sleep) — Lightweight Markdown-only skills for autonomous ML research: cross-model review loops, idea discovery, and experiment automation. No framework, no lock-in — works with Claude Code, Codex, OpenClaw, or any LLM agent.

  • Dse loop

    wanshuiyin/Auto-claude-code-research-in-sleep/skills/skills-codex/dse-loop Skill

    14,663★ repo

    ARIS ⚔️ (Auto-Research-In-Sleep) — Lightweight Markdown-only skills for autonomous ML research: cross-model review loops, idea discovery, and experiment automation. No framework, no lock-in — works with Claude Code, Codex, OpenClaw, or any LLM agent.

  • Embodiment description

    wanshuiyin/Auto-claude-code-research-in-sleep/skills/skills-codex/embodiment-description Skill

    14,663★ repo

    ARIS ⚔️ (Auto-Research-In-Sleep) — Lightweight Markdown-only skills for autonomous ML research: cross-model review loops, idea discovery, and experiment automation. No framework, no lock-in — works with Claude Code, Codex, OpenClaw, or any LLM agent.

  • Exa search

    wanshuiyin/Auto-claude-code-research-in-sleep/skills/skills-codex/exa-search Skill

    14,663★ repo

    ARIS ⚔️ (Auto-Research-In-Sleep) — Lightweight Markdown-only skills for autonomous ML research: cross-model review loops, idea discovery, and experiment automation. No framework, no lock-in — works with Claude Code, Codex, OpenClaw, or any LLM agent.

  • Experiment audit

    wanshuiyin/Auto-claude-code-research-in-sleep/skills/skills-codex/experiment-audit Skill

    14,663★ repo

    ARIS ⚔️ (Auto-Research-In-Sleep) — Lightweight Markdown-only skills for autonomous ML research: cross-model review loops, idea discovery, and experiment automation. No framework, no lock-in — works with Claude Code, Codex, OpenClaw, or any LLM agent.

  • Experiment bridge

    wanshuiyin/Auto-claude-code-research-in-sleep/skills/skills-codex/experiment-bridge Skill

    14,663★ repo

    ARIS ⚔️ (Auto-Research-In-Sleep) — Lightweight Markdown-only skills for autonomous ML research: cross-model review loops, idea discovery, and experiment automation. No framework, no lock-in — works with Claude Code, Codex, OpenClaw, or any LLM agent.

  • Experiment plan

    wanshuiyin/Auto-claude-code-research-in-sleep/skills/skills-codex/experiment-plan Skill

    14,663★ repo

    ARIS ⚔️ (Auto-Research-In-Sleep) — Lightweight Markdown-only skills for autonomous ML research: cross-model review loops, idea discovery, and experiment automation. No framework, no lock-in — works with Claude Code, Codex, OpenClaw, or any LLM agent.

  • Experiment queue

    wanshuiyin/Auto-claude-code-research-in-sleep/skills/skills-codex/experiment-queue Skill

    14,663★ repo

    ARIS ⚔️ (Auto-Research-In-Sleep) — Lightweight Markdown-only skills for autonomous ML research: cross-model review loops, idea discovery, and experiment automation. No framework, no lock-in — works with Claude Code, Codex, OpenClaw, or any LLM agent.

  • Feishu notify

    wanshuiyin/Auto-claude-code-research-in-sleep/skills/skills-codex/feishu-notify Skill

    14,663★ repo

    ARIS ⚔️ (Auto-Research-In-Sleep) — Lightweight Markdown-only skills for autonomous ML research: cross-model review loops, idea discovery, and experiment automation. No framework, no lock-in — works with Claude Code, Codex, OpenClaw, or any LLM agent.

  • Figure description

    wanshuiyin/Auto-claude-code-research-in-sleep/skills/skills-codex/figure-description Skill

    14,663★ repo

    ARIS ⚔️ (Auto-Research-In-Sleep) — Lightweight Markdown-only skills for autonomous ML research: cross-model review loops, idea discovery, and experiment automation. No framework, no lock-in — works with Claude Code, Codex, OpenClaw, or any LLM agent.

  • Figure spec

    wanshuiyin/Auto-claude-code-research-in-sleep/skills/skills-codex/figure-spec Skill

    14,663★ repo

    ARIS ⚔️ (Auto-Research-In-Sleep) — Lightweight Markdown-only skills for autonomous ML research: cross-model review loops, idea discovery, and experiment automation. No framework, no lock-in — works with Claude Code, Codex, OpenClaw, or any LLM agent.

  • Formula derivation

    wanshuiyin/Auto-claude-code-research-in-sleep/skills/skills-codex/formula-derivation Skill

    14,663★ repo

    ARIS ⚔️ (Auto-Research-In-Sleep) — Lightweight Markdown-only skills for autonomous ML research: cross-model review loops, idea discovery, and experiment automation. No framework, no lock-in — works with Claude Code, Codex, OpenClaw, or any LLM agent.

  • Gemini search

    wanshuiyin/Auto-claude-code-research-in-sleep/skills/skills-codex/gemini-search Skill

    14,663★ repo

    ARIS ⚔️ (Auto-Research-In-Sleep) — Lightweight Markdown-only skills for autonomous ML research: cross-model review loops, idea discovery, and experiment automation. No framework, no lock-in — works with Claude Code, Codex, OpenClaw, or any LLM agent.

  • Grant proposal

    wanshuiyin/Auto-claude-code-research-in-sleep/skills/skills-codex/grant-proposal Skill

    14,663★ repo

    ARIS ⚔️ (Auto-Research-In-Sleep) — Lightweight Markdown-only skills for autonomous ML research: cross-model review loops, idea discovery, and experiment automation. No framework, no lock-in — works with Claude Code, Codex, OpenClaw, or any LLM agent.

  • Idea creator

    wanshuiyin/Auto-claude-code-research-in-sleep/skills/skills-codex/idea-creator Skill

    14,663★ repo

    ARIS ⚔️ (Auto-Research-In-Sleep) — Lightweight Markdown-only skills for autonomous ML research: cross-model review loops, idea discovery, and experiment automation. No framework, no lock-in — works with Claude Code, Codex, OpenClaw, or any LLM agent.

  • Idea discovery robot

    wanshuiyin/Auto-claude-code-research-in-sleep/skills/skills-codex/idea-discovery-robot Skill

    14,663★ repo

    ARIS ⚔️ (Auto-Research-In-Sleep) — Lightweight Markdown-only skills for autonomous ML research: cross-model review loops, idea discovery, and experiment automation. No framework, no lock-in — works with Claude Code, Codex, OpenClaw, or any LLM agent.

  • Idea discovery

    wanshuiyin/Auto-claude-code-research-in-sleep/skills/skills-codex/idea-discovery Skill

    14,663★ repo

    ARIS ⚔️ (Auto-Research-In-Sleep) — Lightweight Markdown-only skills for autonomous ML research: cross-model review loops, idea discovery, and experiment automation. No framework, no lock-in — works with Claude Code, Codex, OpenClaw, or any LLM agent.

  • Integrity forensics

    wanshuiyin/Auto-claude-code-research-in-sleep/skills/skills-codex/integrity-forensics Skill

    14,663★ repo

    ARIS ⚔️ (Auto-Research-In-Sleep) — Lightweight Markdown-only skills for autonomous ML research: cross-model review loops, idea discovery, and experiment automation. No framework, no lock-in — works with Claude Code, Codex, OpenClaw, or any LLM agent.

  • Interview cheatsheet

    wanshuiyin/Auto-claude-code-research-in-sleep/skills/skills-codex/interview-cheatsheet Skill

    14,663★ repo

    ARIS ⚔️ (Auto-Research-In-Sleep) — Lightweight Markdown-only skills for autonomous ML research: cross-model review loops, idea discovery, and experiment automation. No framework, no lock-in — works with Claude Code, Codex, OpenClaw, or any LLM agent.

  • Invention structuring

    wanshuiyin/Auto-claude-code-research-in-sleep/skills/skills-codex/invention-structuring Skill

    14,663★ repo

    ARIS ⚔️ (Auto-Research-In-Sleep) — Lightweight Markdown-only skills for autonomous ML research: cross-model review loops, idea discovery, and experiment automation. No framework, no lock-in — works with Claude Code, Codex, OpenClaw, or any LLM agent.

  • Jurisdiction format

    wanshuiyin/Auto-claude-code-research-in-sleep/skills/skills-codex/jurisdiction-format Skill

    14,663★ repo

    ARIS ⚔️ (Auto-Research-In-Sleep) — Lightweight Markdown-only skills for autonomous ML research: cross-model review loops, idea discovery, and experiment automation. No framework, no lock-in — works with Claude Code, Codex, OpenClaw, or any LLM agent.

  • Kill argument

    wanshuiyin/Auto-claude-code-research-in-sleep/skills/skills-codex/kill-argument Skill

    14,663★ repo

    ARIS ⚔️ (Auto-Research-In-Sleep) — Lightweight Markdown-only skills for autonomous ML research: cross-model review loops, idea discovery, and experiment automation. No framework, no lock-in — works with Claude Code, Codex, OpenClaw, or any LLM agent.

  • Mermaid diagram

    wanshuiyin/Auto-claude-code-research-in-sleep/skills/skills-codex/mermaid-diagram Skill

    14,663★ repo

    ARIS ⚔️ (Auto-Research-In-Sleep) — Lightweight Markdown-only skills for autonomous ML research: cross-model review loops, idea discovery, and experiment automation. No framework, no lock-in — works with Claude Code, Codex, OpenClaw, or any LLM agent.

  • Meta apply

    wanshuiyin/Auto-claude-code-research-in-sleep/skills/skills-codex/meta-apply Skill

    14,663★ repo

    ARIS ⚔️ (Auto-Research-In-Sleep) — Lightweight Markdown-only skills for autonomous ML research: cross-model review loops, idea discovery, and experiment automation. No framework, no lock-in — works with Claude Code, Codex, OpenClaw, or any LLM agent.

  • Meta optimize

    wanshuiyin/Auto-claude-code-research-in-sleep/skills/skills-codex/meta-optimize Skill

    14,663★ repo

    ARIS ⚔️ (Auto-Research-In-Sleep) — Lightweight Markdown-only skills for autonomous ML research: cross-model review loops, idea discovery, and experiment automation. No framework, no lock-in — works with Claude Code, Codex, OpenClaw, or any LLM agent.

  • Monitor experiment

    wanshuiyin/Auto-claude-code-research-in-sleep/skills/skills-codex/monitor-experiment Skill

    14,663★ repo

    ARIS ⚔️ (Auto-Research-In-Sleep) — Lightweight Markdown-only skills for autonomous ML research: cross-model review loops, idea discovery, and experiment automation. No framework, no lock-in — works with Claude Code, Codex, OpenClaw, or any LLM agent.

  • Novelty check

    wanshuiyin/Auto-claude-code-research-in-sleep/skills/skills-codex/novelty-check Skill

    14,663★ repo

    ARIS ⚔️ (Auto-Research-In-Sleep) — Lightweight Markdown-only skills for autonomous ML research: cross-model review loops, idea discovery, and experiment automation. No framework, no lock-in — works with Claude Code, Codex, OpenClaw, or any LLM agent.

  • Openalex

    wanshuiyin/Auto-claude-code-research-in-sleep/skills/skills-codex/openalex Skill

    14,663★ repo

    ARIS ⚔️ (Auto-Research-In-Sleep) — Lightweight Markdown-only skills for autonomous ML research: cross-model review loops, idea discovery, and experiment automation. No framework, no lock-in — works with Claude Code, Codex, OpenClaw, or any LLM agent.

  • Overleaf sync

    wanshuiyin/Auto-claude-code-research-in-sleep/skills/skills-codex/overleaf-sync Skill

    14,663★ repo

    ARIS ⚔️ (Auto-Research-In-Sleep) — Lightweight Markdown-only skills for autonomous ML research: cross-model review loops, idea discovery, and experiment automation. No framework, no lock-in — works with Claude Code, Codex, OpenClaw, or any LLM agent.

  • Paper claim audit

    wanshuiyin/Auto-claude-code-research-in-sleep/skills/skills-codex/paper-claim-audit Skill

    14,663★ repo

    ARIS ⚔️ (Auto-Research-In-Sleep) — Lightweight Markdown-only skills for autonomous ML research: cross-model review loops, idea discovery, and experiment automation. No framework, no lock-in — works with Claude Code, Codex, OpenClaw, or any LLM agent.

  • Paper compile

    wanshuiyin/Auto-claude-code-research-in-sleep/skills/skills-codex/paper-compile Skill

    14,663★ repo

    ARIS ⚔️ (Auto-Research-In-Sleep) — Lightweight Markdown-only skills for autonomous ML research: cross-model review loops, idea discovery, and experiment automation. No framework, no lock-in — works with Claude Code, Codex, OpenClaw, or any LLM agent.

  • Paper figure

    wanshuiyin/Auto-claude-code-research-in-sleep/skills/skills-codex/paper-figure Skill

    14,663★ repo

    ARIS ⚔️ (Auto-Research-In-Sleep) — Lightweight Markdown-only skills for autonomous ML research: cross-model review loops, idea discovery, and experiment automation. No framework, no lock-in — works with Claude Code, Codex, OpenClaw, or any LLM agent.

  • Paper illustration image2

    wanshuiyin/Auto-claude-code-research-in-sleep/skills/skills-codex/paper-illustration-image2 Skill

    14,663★ repo

    ARIS ⚔️ (Auto-Research-In-Sleep) — Lightweight Markdown-only skills for autonomous ML research: cross-model review loops, idea discovery, and experiment automation. No framework, no lock-in — works with Claude Code, Codex, OpenClaw, or any LLM agent.

  • Paper illustration

    wanshuiyin/Auto-claude-code-research-in-sleep/skills/skills-codex/paper-illustration Skill

    14,663★ repo

    ARIS ⚔️ (Auto-Research-In-Sleep) — Lightweight Markdown-only skills for autonomous ML research: cross-model review loops, idea discovery, and experiment automation. No framework, no lock-in — works with Claude Code, Codex, OpenClaw, or any LLM agent.

  • Paper plan

    wanshuiyin/Auto-claude-code-research-in-sleep/skills/skills-codex/paper-plan Skill

    14,663★ repo

    ARIS ⚔️ (Auto-Research-In-Sleep) — Lightweight Markdown-only skills for autonomous ML research: cross-model review loops, idea discovery, and experiment automation. No framework, no lock-in — works with Claude Code, Codex, OpenClaw, or any LLM agent.

  • Paper poster html

    wanshuiyin/Auto-claude-code-research-in-sleep/skills/skills-codex/paper-poster-html Skill

    14,663★ repo

    ARIS ⚔️ (Auto-Research-In-Sleep) — Lightweight Markdown-only skills for autonomous ML research: cross-model review loops, idea discovery, and experiment automation. No framework, no lock-in — works with Claude Code, Codex, OpenClaw, or any LLM agent.

  • Paper poster

    wanshuiyin/Auto-claude-code-research-in-sleep/skills/skills-codex/paper-poster Skill

    14,663★ repo

    ARIS ⚔️ (Auto-Research-In-Sleep) — Lightweight Markdown-only skills for autonomous ML research: cross-model review loops, idea discovery, and experiment automation. No framework, no lock-in — works with Claude Code, Codex, OpenClaw, or any LLM agent.

  • Paper slides

    wanshuiyin/Auto-claude-code-research-in-sleep/skills/skills-codex/paper-slides Skill

    14,663★ repo

    ARIS ⚔️ (Auto-Research-In-Sleep) — Lightweight Markdown-only skills for autonomous ML research: cross-model review loops, idea discovery, and experiment automation. No framework, no lock-in — works with Claude Code, Codex, OpenClaw, or any LLM agent.

  • Paper talk

    wanshuiyin/Auto-claude-code-research-in-sleep/skills/skills-codex/paper-talk Skill

    14,663★ repo

    ARIS ⚔️ (Auto-Research-In-Sleep) — Lightweight Markdown-only skills for autonomous ML research: cross-model review loops, idea discovery, and experiment automation. No framework, no lock-in — works with Claude Code, Codex, OpenClaw, or any LLM agent.

  • Paper write

    wanshuiyin/Auto-claude-code-research-in-sleep/skills/skills-codex/paper-write Skill

    14,663★ repo

    ARIS ⚔️ (Auto-Research-In-Sleep) — Lightweight Markdown-only skills for autonomous ML research: cross-model review loops, idea discovery, and experiment automation. No framework, no lock-in — works with Claude Code, Codex, OpenClaw, or any LLM agent.

  • Paper writing

    wanshuiyin/Auto-claude-code-research-in-sleep/skills/skills-codex/paper-writing Skill

    14,663★ repo

    ARIS ⚔️ (Auto-Research-In-Sleep) — Lightweight Markdown-only skills for autonomous ML research: cross-model review loops, idea discovery, and experiment automation. No framework, no lock-in — works with Claude Code, Codex, OpenClaw, or any LLM agent.

  • Patent novelty check

    wanshuiyin/Auto-claude-code-research-in-sleep/skills/skills-codex/patent-novelty-check Skill

    14,663★ repo

    ARIS ⚔️ (Auto-Research-In-Sleep) — Lightweight Markdown-only skills for autonomous ML research: cross-model review loops, idea discovery, and experiment automation. No framework, no lock-in — works with Claude Code, Codex, OpenClaw, or any LLM agent.

  • Patent pipeline

    wanshuiyin/Auto-claude-code-research-in-sleep/skills/skills-codex/patent-pipeline Skill

    14,663★ repo

    ARIS ⚔️ (Auto-Research-In-Sleep) — Lightweight Markdown-only skills for autonomous ML research: cross-model review loops, idea discovery, and experiment automation. No framework, no lock-in — works with Claude Code, Codex, OpenClaw, or any LLM agent.

  • Patent review

    wanshuiyin/Auto-claude-code-research-in-sleep/skills/skills-codex/patent-review Skill

    14,663★ repo

    ARIS ⚔️ (Auto-Research-In-Sleep) — Lightweight Markdown-only skills for autonomous ML research: cross-model review loops, idea discovery, and experiment automation. No framework, no lock-in — works with Claude Code, Codex, OpenClaw, or any LLM agent.

  • Pixel art

    wanshuiyin/Auto-claude-code-research-in-sleep/skills/skills-codex/pixel-art Skill

    14,663★ repo

    ARIS ⚔️ (Auto-Research-In-Sleep) — Lightweight Markdown-only skills for autonomous ML research: cross-model review loops, idea discovery, and experiment automation. No framework, no lock-in — works with Claude Code, Codex, OpenClaw, or any LLM agent.

  • Prior art search

    wanshuiyin/Auto-claude-code-research-in-sleep/skills/skills-codex/prior-art-search Skill

    14,663★ repo

    ARIS ⚔️ (Auto-Research-In-Sleep) — Lightweight Markdown-only skills for autonomous ML research: cross-model review loops, idea discovery, and experiment automation. No framework, no lock-in — works with Claude Code, Codex, OpenClaw, or any LLM agent.