ARIS
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.
What is ARIS?
ARIS is a Model Context Protocol (MCP) server that allows AI assistants like Claude, Cursor, and VS Code to 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 wi...
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.
This server falls under the Coding Agents category on MCPgee, the world's largest MCP server directory with 33,000+ servers.
Features
- ARIS ⚔️ (Auto-Research-In-Sleep) — Lightweight Markdown-only
Use Cases
Maintainer
Works with
Installation
Manual Installation
npx auto-claude-code-research-in-sleepConfiguration
Configuration Details
claude_desktop_config.json
Performance
Response Metrics
Resource Usage
How to Set Up and Use ARIS
ARIS (Auto-Research-In-Sleep) is a collection of over 79 composable Markdown-only skills for autonomous machine learning research that works with Claude Code, Codex, OpenClaw, and other LLM agents. It orchestrates complete research pipelines — from literature search and novelty verification through experiment execution on local or remote GPUs to paper writing, peer-review response generation, and conference talk preparation. Because it uses no framework or library lock-in, the skills drop directly into any agent's skills directory and chain together through inline configuration parameters.
Prerequisites
- Claude Code, Codex, or another LLM agent that supports a skills directory
- Git to clone the repository
- An EXA_API_KEY for web search capabilities (optional but recommended)
- A GEMINI_API_KEY for AI-generated paper illustrations (optional)
- ARIS_VERIFY_EMAIL set for CrossRef academic API polite pool access (optional)
Clone the ARIS repository
Clone the repository to get the full skill set and installer tools.
git clone https://github.com/wanshuiyin/Auto-claude-code-research-in-sleep.gitInstall skills into your project (recommended)
Run the install script pointing to your project directory. This creates a local symlink so updates are easy to pull in.
bash Auto-claude-code-research-in-sleep/tools/install_aris.sh ~/your-projectInstall skills globally (alternative)
Copy the skills directly into Claude Code's global skills directory to make them available across all projects.
cp -r Auto-claude-code-research-in-sleep/skills/* ~/.claude/skills/Set optional environment variables
Export the API keys that unlock additional capabilities. Add these to your shell profile or CLAUDE.md for persistence.
export EXA_API_KEY=your-exa-api-key
export GEMINI_API_KEY=your-gemini-api-key
export [email protected]Configure a remote GPU (optional)
For experiment execution on a remote server, add GPU configuration to your project's CLAUDE.md file. ARIS supports local GPUs, SSH remote servers, and on-demand Vast.ai provisioning.
# Add to CLAUDE.md:
gpu: remote
gpu_host: [email protected]
gpu_port: 22
gpu_workspace: /path/to/workspace
conda_hook: source ~/miniconda3/bin/activate researchRun your first research pipeline
Open Claude Code and invoke the /research-pipeline skill with a research topic. Append inline overrides to control effort level, reviewer model, and other parameters.
ARIS Examples
Client configuration
ARIS installs as agent skills, not a traditional MCP server. This shows how to configure Codex as a cross-model reviewer via MCP alongside ARIS.
{
"mcpServers": {
"codex": {
"command": "codex",
"args": ["mcp-server"]
}
}
}Prompts to try
Invoke ARIS skills in Claude Code using these prompts. Append parameters with double-dash syntax.
- "/research-pipeline 'factorized gap in discrete diffusion LMs'"
- "/research-pipeline 'improve method X' -- ref paper: https://arxiv.org/abs/2406.04329"
- "/rebuttal 'paper/ + reviews' -- venue: ICML, character limit: 5000"
- "/paper-talk 'paper/' -- venue: ICLR, assurance: conference-ready"
- "/exa-search 'latest advances in diffusion models 2024'"Troubleshooting ARIS
Skills not found after installation
Verify the skills were copied to the correct directory for your agent (e.g., ~/.claude/skills/ for Claude Code). Run 'ls ~/.claude/skills/' and confirm the ARIS skill files are present. Reload your agent after installing.
Research pipeline stalls waiting for reviewer
The default reviewer is Codex (GPT-5.5). If Codex is unavailable, switch to manual review mode: append '-- reviewer: manual' to your pipeline command. This pauses for you to paste the review into any model.
EXA search or paper fetch fails with API errors
Confirm EXA_API_KEY is exported in your shell (echo $EXA_API_KEY). If ARIS_STREAM_IDLE_TIMEOUT_SECS is too short for slow connections, increase it: export ARIS_STREAM_IDLE_TIMEOUT_SECS=300
Frequently Asked Questions about ARIS
What is ARIS?
ARIS is a Model Context Protocol (MCP) server that 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. It connects AI assistants to external tools and data sources through a standardized interface.
How do I install ARIS?
Follow the installation instructions on the ARIS GitHub repository. Clone the repo, install dependencies, and add the server config to your AI client.
Which AI clients work with ARIS?
ARIS works with all major MCP-compatible AI clients including Claude Desktop, Claude Code, Cursor, VS Code (GitHub Copilot), Windsurf, and Cline.
Is ARIS free to use?
Yes, ARIS is open source and available under the MIT license. You can use it freely in both personal and commercial projects.
ARIS Alternatives — Similar Coding Agents Servers
Looking for alternatives to ARIS? Here are other popular coding agents servers you can use with Claude, Cursor, and VS Code.
Dify
★ 142.2kProduction-ready platform for agentic workflow development.
Ruflo
★ 54.0k🌊 The leading agent orchestration platform for Claude. Deploy intelligent multi-agent swarms, coordinate autonomous workflows, and build conversational AI systems. Features enterprise-grade architecture, self-learning swarm intelligence, RAG integrat
Goose
★ 45.7kan open source, extensible AI agent that goes beyond code suggestions - install, execute, edit, and test with any LLM
Antigravity Awesome Skills
★ 38.3kInstallable GitHub library of 1,400+ agentic skills for Claude Code, Cursor, Codex CLI, Gemini CLI, Antigravity, and more. Includes installer CLI, bundles, workflows, and official/community skill collections.
AgentScope
★ 25.5kBuild and run agents you can see, understand and trust.
Serena
★ 24.5kA coding agent toolkit that provides IDE-like semantic code retrieval and editing tools, enabling LLMs to efficiently navigate and modify codebases using symbol-level operations instead of basic file reading and string replacements.
Browse More Coding Agents MCP Servers
Explore all coding agents servers available in the MCPgee directory. Each server includes setup guides for Claude, Cursor, and VS Code.
Set Up ARIS in Your Editor
Choose your AI client for step-by-step setup instructions.
Quick Config Preview
Add this to your claude_desktop_config.json or .cursor/mcp.json
Ready to use ARIS?
Browse our complete directory of 33,000+ MCP servers, read setup guides for your editor, and start building with the Model Context Protocol.