#5151st of 73 in Skills, plugins and rules
auto-claude-code-research-in-sleep
Skill-based research workflow for Claude Code with cross-model review
Documentation ↗
(opens in a new tab)Website ↗
(opens in a new tab)Repository on GitHub ↗
(opens in a new tab)
- Stars
- 17.5k
- License
- MIT
- Last commit
- Oct 2026
- Last release
- Sep 2026
- Language
- Python
3,969 on 25 Mar 202617,483 on 14 Oct 2026
From GitHub's daily star counts.
Checked on GitHub.
Overview
ARIS is a set of 82 skills that run an autonomous research workflow (idea discovery, experiments, paper writing) inside Claude Code, Codex CLI, Cursor and similar agents. An executor model drives the work while an independent reviewer, Codex through an MCP bridge over `codex exec` by default, critiques it. It also ships as a Claude Code and Codex plugin and as a standalone Rust CLI, ARIS-Code.
Who it is for: Researchers using Claude Code or Codex to automate experiments and paper writing
Strengths
- Works across Claude Code, Codex CLI, Cursor, Trae, Copilot CLI and OpenClaw
- Separate reviewer model checks the executor's output; alternative model combinations are documented
- Ships as plain skills, plugins, or a standalone CLI with prebuilt releases
- MIT licensed, with frequent releases (ARIS-Code v0.4.28 on 2026-09-28)
Weaknesses
- Default reviewer depends on codex-cli; its MCP entry point was removed in 0.154
- Aimed at research and paper workflows, not general application coding
- Large surface (82 skills, many helpers) and a changelog-heavy README
- Per-host adaptations live in separate docs, so support varies by agent
What it needs
- Needs Claude Code or another supported agent host, codex-cli (default reviewer), Python
- Models: Claude Opus 5, Claude Fable 5.1, GPT-6-Astra, ModelScope-hosted models, DeepSeek
Also in Skills, plugins and rules
See all 73| Rank | Project | Score |
|---|---|---|
| 1 | skillsAgent skills for planning, TDD, debugging and code review | 94 out of 100 |
| 2 | eccAgents, skills, hooks and rules bundle for Claude Code, Codex and other harnesses | 94 out of 100 |
| 3 | cavemanCuts token use in coding agents with terse replies and compressed inputs | 93 out of 100 |
| #4 | ponytailSkill prompt that makes coding agents write less code and add tests | 92 out of 100 |
| #5 | ui-ux-pro-max-skillDesign-system skill that gives coding agents searchable UI styles, palettes and fonts | 91 out of 100 |