AgentIndex icon
AgentIndex
ToolsMCP ServersHOTCategoriesTrendingNewBest ListsCompareGuides
Submit Tool
ToolsMCP ServersHOTCategoriesTrendingNewBest ListsCompareGuides
Home/
Workflow Automation/
Auto-claude-code-research-in-sleep
Auto-claude-code-research-in-sleep logo

Auto-claude-code-research-in-sleep

Active·★ 19.4k·MIT·Updated 2026-11-17
★ Most Popular★ Trending

ARIS (Auto-Research-In-Sleep) is a set of custom Claude Code skills for autonomous machine-learning research. The idea is cross-model collaboration: Claude Code drives the research while an external LLM, by default GPT-5.4 at extra-high reasoning reached through Codex MCP, acts as the reviewer. The README argues that a single model reviewing its own work tends to fall into local minima and that two complementary models, fast execution plus rigorous critique, do better.

It offers 18 composable skills chained into workflows: idea discovery (literature survey, brainstorming 8 to 12 ideas, novelty checks, GPU pilot experiments, a ranked report), an auto review loop of four autonomous review rounds, a paper writing pipeline from outline to figures, LaTeX and PDF, peer review of other papers, and GPU deployment with rsync, screen sessions and multi-GPU runs. Human-in-the-loop checkpoints are configurable (AUTO_PROCEED), model combinations beyond Claude and GPT are supported, and optional Zotero, Obsidian and Feishu/Lark integrations exist. Install by copying the skills into ~/.claude/skills and adding the Codex MCP server.

This write-up is based on the project's README and public repository data. We have not hand-tested the tool. Data as of 2026-11-07.

License
MIT
Latest release
v0.4.28 · 2026-09-28
Last commit
2026-11-06
Contributors
106
Forks
1,531
Open issues and PRs
75
First published
2026-03-10

Source: GitHub, npm and PyPI public APIs.

#ai-research#ai-tools#aris#autonomous-agent#claude#claude-code#claude-code-skills#codex
$ Install
$ git clone https://github.com/wanshuiyin/Auto-claude-code-research-in-sleep.git && cp -r Auto-claude-code-research-in-sleep/skills/* ~/.claude/skills/ && npm install -g @openai/codex && claude mcp add codex -s user -- codex mcp-server
↗ Visit site★ GitHub
01

Features

0118 composable skills for flexible workflow chaining.
02Automated idea discovery including literature survey, brainstorming, novelty check, and GPU pilot experiments.
03Autonomous multi-round review loop to iteratively improve research with experiments.
04Comprehensive paper writing pipeline from narrative to submission-ready LaTeX/PDF.
05Cross-model collaboration for adversarial review, breaking single-model blind spots.
02

Why choose it

+Covers the research cycle end to end: literature, ideas, experiments, review loops and paper writing.
+Cross-model design uses a second model as an adversarial reviewer instead of Claude reviewing itself.
+Human-in-the-loop checkpoints are configurable, from step-by-step approval to full autopilot.
+Supports alternative model pairs (such as GLM with GPT or MiniMax) and optional Zotero, Obsidian and Feishu/Lark integration.
03

Trade-offs

!The default setup needs both Claude Code and an OpenAI Codex setup for the reviewer, which means two paid services and extra token cost.
!The score improvements shown (for example 5 to 7.5 out of 10 overnight) come from the project's own runs and a model-based reviewer, not an independent evaluation.
!Autonomous experiments run on your GPU machines and write papers unattended, so results need human verification before being trusted.
05

Maintenance check

ActiveChecked: 2026-11-06
Last commit
2026-11-06
Commits, last 90 days
100+
Releases, last 90 days
7
Latest release
v0.4.28 (2026-09-28)
Open issues
75
Recent issues with at least one comment
7/11
Commits per week, last 12 weeks
Commits per week, last 12 weeks: 16, 19, 4, 13, 37, 0, 12, 7, 16, 23, 2, 9

Oldest week on the left; the latest four weeks are orange. Source: GitHub commit activity.

Status rule: archived on GitHub = Archived; last commit within 90 days = Active; 91 to 365 days = Slowing; older = Stale. All figures come from the GitHub API on the date shown.

Most discussed open issues

  • #115 Code版本Windows平台25 comments · 2026-04-07
  • #240 README写的太繁杂了,可读性差,希望精简流程,增加标题层级编号6 comments · 2026-05-22
  • #430 严重bug,停不下来4 comments · 2026-09-03
  • #111 请问独立安装版相较于普通版有什么优势?4 comments · 2026-04-05
  • #433 [Codex] Mark workflow-only skills as explicit-only to reduce routing/context overhead3 comments · 2026-09-07

Issue titles are quoted from GitHub. Open the link to check whether a problem still applies to the version you use.

04

Who it fits

Pick it if
  • You use Claude Code and want an autonomous skill loop for machine learning research that ideates, executes experiments, and reviews results overnight.
  • You want a dual-model peer-review mechanism where a secondary LLM audits research ideas and code before running.
Look elsewhere if
  • You need an autonomous web-search deep research agent for general factual reports; GPT Researcher is tailored for text reports.
  • You do not use Claude Code CLI or prefer a graphical node-based workflow builder.
06

How it differs

GPT Researcher logovs GPT Researcher★ 30.8k

Auto-claude-code-research-in-sleep automates iterative ML coding, training scripts, and experiment loops via Claude Code. GPT Researcher focuses on web research, source aggregation, and structured factual reporting.

Full comparison →
07

Compatibility

Claude Code
Core Executor
Verified via docs
Codex CLI
Critical Reviewer
Verified via docs
LaTeX
Paper Compilation
Verified via docs
GLM
Alternative Executor
Verified via docs
MiniMax
Alternative Reviewer
Verified via docs
Zotero
Literature Search
Verified via docs
08

Quick start

1
$ git clone https://github.com/wanshuiyin/Auto-claude-code-research-in-sleep.git
2
$ cp -r Auto-claude-code-research-in-sleep/skills/* ~/.claude/skills/
3
$ npm install -g @openai/codex
4
$ claude mcp add codex -s user -- codex mcp-server
09

Use cases

↳Explore new research areas and discover novel ideas through literature surveys and brainstorming.
↳Automate the iterative review and refinement of research projects, including running experiments, until submission-ready.
↳Transform research narratives into submission-ready academic papers with a full writing and formatting pipeline.
10

Alternatives

⚡Top VS:
VSGPT Researcher★30.8k→VSche-ical-mcp★45→VSclaude-code-tips★30→VSapra-fleet★107→
GPT Researcher logo
GPT Researcher★ 30.8k
An autonomous agent that conducts deep research on any data using any LLM provider, producing detailed reports with citations from web and local sources.
VSCompare→
zabbix-mcp-server logo
zabbix-mcp-server★ 254
MCP server for the complete Zabbix API — 220 tools, multi-server support, bearer auth, systemd ready. Works with ChatGPT, Claude, VS Code, Codex, JetBrains and any MCP client.
che-ical-mcp logo
che-ical-mcp★ 45
macOS Calendar & Reminders MCP server with 28 tools - Native Swift EventKit integration for Claude AI. Attendee/organizer info, per-event timezone, undo/redo, --cli mode, --setup TCC, batch ops. Supports iCloud, Google, Exchange.
VSCompare→
claude-code-tips logo
claude-code-tips★ 30
practical patterns for agentic coding: hooks, agents, automation. built from hundreds of claude code sessions
VSCompare→
notebooklm-skill logo
notebooklm-skill★ 502
NotebookLM does the research, Claude writes the content. Research → Synthesis → Content Creation → Publishing. Claude Code Skill + MCP Server.
apra-fleet logo
apra-fleet★ 107
AI agents that write code, review each other's work, and coordinate across your machines
VSCompare→
NyxID logo
NyxID★ 41
Connect AI agents to any API, anywhere. Securely. Open-source gateway that proxies requests, injects credentials automatically, punches through NAT to reach localhost services, and wraps REST APIs as MCP tools. Per-agent isolation. Never expose a raw key.
VSCompare→
whetstone logo
whetstone★ 37
AI-powered development tools. 19 agents, 22 commands, 30 skills, 1 hook, 1 MCP server for code review, research, design, and workflow automation.
VSCompare→
See all alternatives →

Comments

Log in to leave a comment

No comments yet. Be the first!

On this page
01Features02Why choose it03Trade-offs05Maintenance check04Who it fits06How it differs07Compatibility08Quick start09Use cases10Alternatives
wanshuiyin /
Auto-claude-code-research-in-sleep
↗

Claude Code skills that run ML research loops with a second model as reviewer

19.4k1.5kMIT
Last commit1d ago
StatusActive
CategoryWorkflow Automation
Contributors106
Latest releasev0.4.28
Links
Documentation↗Discussion↗Issues↗Releases↗
Cloud Infrastructure
Ad

Deploy on DigitalOcean

Host your MCP servers, Docker containers, and AI agents with $200 free credit over 60 days.

Claim $200 Free Credit
© 2026 AgentIndex.app|Built by a 10-year iOS Developer.
Best by use caseAboutContactAdvertisePrivacy PolicyTerms of UseQYSGitHubBuy me a coffee ☕

Browse by Category

MCP ServersCode AssistantWorkflow AutomationRAG / Knowledge BaseMulti-AgentBrowser AutomationLLM InfraDev ToolingObservability

Not affiliated with Anthropic, OpenAI or Microsoft.