wanshuiyin/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.

ARIS (Auto-claude-code-research-in-sleep)

A methodology + toolkit that lets you hand a research task to an AI coding agent (the executor), while a second, independent model (the reviewer) checks every step. It's designed to run long, multi-step research automatically — you can start it before bed and have an audited result by morning.

The project ships as:

  • Skills: reusable instruction sets (prompt templates and scripts) that plug into Claude Code, Codex CLI, Cursor, Trae, GitHub Copilot CLI, and others.
  • ARIS-Code: a standalone command-line tool (Rust-based) that bundles 83 skills, manages sessions, tracks costs, and lets you resume.
  • Plugins for Claude Code and Codex CLI, and a DeepSeek Harness variant.

Key idea: the executor does the work, but a separate reviewer model (by default Codex via MCP) verifies each step for correctness, fabrication, and missed edge cases. This prevents the classic problem of an AI confidently hallucinating results.

The project also includes a macOS status widget (ARIS-Monitor) and a family of related tools (ARIS-in-AI-Offer for interview prep, HERO to prevent over-engineering, Anti-Autoresearch to detect fabricated AI outputs).

The README also references a technical report (arXiv:2605.03042) and highlights it as Hugging Face's #1 Paper of the Day, so it's a peer-reviewed (or at least community-recognized) methodology, not just a hobby script.

Related

  • Project
    chaseai-yt/claudex-loop

    Claudex Loop is a multi-agent workflow that improves AI coding reliability by requiring independent AI models to review plans and inspect code generated by a primary builder.

  • Project
    OpenNSWM-Lab/FAROS

    FAROS is a multi-agent AI system designed to automate scientific research workflows, from literature review and experiment planning to evidence auditing and paper writing, while ensuring all findings are traceable and verifiable.

  • Project
    EvoMap/AutoResearch

    An open-source agent workflow for AI/ML research that automates idea generation and experiment execution to produce paper-ready evidence.

  • Project
    Imbad0202/academic-research-skills-codex

    ARS‑Codex is a Codex‑compatible skill that bundles the Academic Research Skills workflow suite (originally built for Claude Code) into a single plug‑in. After installing the plugin, users can invoke `$academic‑research‑suite` and let Codex route the request to one of six research‑oriented workflows—deep research, paper drafting, reviewer simulation, full research‑to‑paper pipeline, systematic‑review screener, or experiment planning. The repo contains the skill manifest, prompt templates, optional runtime adapters, and installation scripts; it does not include a new model.

  • Project
    Weizhena/Deep-Research-skills

    A structured research workflow for Claude Code, OpenCode, and Codex that enables systematic outline generation, deep web investigation, and automated report creation.