Auto-claude-code-research-in-sleep
ARIS ⚔️ (Auto-Research-In-Sleep) — Lightweight Markdown-only skills for autonomous ML research: cross-model review loops, idea discovery, and experiment automat
What is Auto-claude-code-research-in-sleep?
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.
Overview of Auto-claude-code-research-in-sleep
What it is
ARIS ⚔️ (Auto-Research-In-Sleep) is a lightweight Markdown-only methodology for autonomous machine learning research. It provides skills for cross-model review loops, idea discovery, and experiment automation. ARIS is not a platform or framework; it works with Claude Code, Codex, OpenClaw, or any LLM agent without lock-in. It is open source under the MIT license.
Key features
- Cross-model review loops: an executor drives experiments while an independent model (Codex MCP by default) reviews results.
- Markdown-only skills: no framework, no dependencies; skills are plain Markdown files installed directly.
- Standalone CLI and plugins: ARIS also ships as a CLI and as plugins for various environments.
- Multi-model support: alternative model combinations, including ModelScope-hosted models, can be used; no Claude or OpenAI API required.
- Per-thread resume and
ultraaudits: supports resuming interrupted sessions and running deep audits with extended budgets.
Use cases
- Autonomous ML research: running experiments and review cycles without human intervention.
- Idea discovery: generating and validating research ideas through cross-model critique.
- Experiment automation: scripting and automating experiment pipelines across different LLM agents.
- Interview preparation: companion project EasyAIOffer provides 34 bilingual ML/LLM interview cheat sheets.
Getting started
See the repository README for installation and quick-start instructions.
License and pricing
Licensed under the MIT License. Pricing is open source (free).
Based on the project's public README and repository metadata.