ARIS: Lightweight Markdown-Driven Automated ML Research Tool with Cross-Model Review Loops

ARIS (Auto-Research-In-Sleep) is a lightweight, Markdown-only skill set for autonomous ML research. Core features include cross-model review loops, idea discovery, and experiment automation. It requires no framework or vendor lock-in and works with any LLM agent (e.g., Claude Code, Codex, OpenClaw). Users write skill descriptions in Markdown to drive agents in tasks like paper review, hypothesis generation, and experiment execution. For developers automating research with Claude Code or Codex, ARIS offers a minimal alternative—no complex SDKs or APIs to learn. With 13,016 GitHub stars and 190 new stars in the last 7 days, community interest is high. Teams building autonomous research pipelines without over-engineering will find ARIS easy to integrate into existing workflows.