ARIS (Auto-Research-In-Sleep)
β Text source-checked Nov 18, 2026Lightweight Markdown-only skills for autonomous ML research: cross-model review loops, idea discovery, and experiment automation.
Source-checked against the vendor's site Β· rebuilt from the vendor's own pagesProfile updated Link checked How we rate
In short
ARIS is a free, MIT-licensed methodology for autonomous ML research using Markdown skills. It enables cross-model review loops and experiment automation across various CLI tools.
ARIS is a methodology rather than a platform, focusing on the research workflow itself. ARIS functions as a skill-based workflow compatible with Claude Code, Codex CLI, Cursor, Trae, Antigravity, GitHub Copilot CLI, OpenClaw, and DeepSeek Harness.
The architecture involves an executor model driving the process while an independent model, such as Codex MCP by default, performs reviews. ARIS supports alternative model combinations, including those hosted on ModelScope, which do not require Claude or OpenAI API keys.
ARIS is available as a standalone CLI and as plugins in addition to direct skill installations.
According to ARIS (Auto-Research-In-Sleep): ARIS is a methodology rather than a platform, Focusing on the research workflow itself. Its pages list this strength: Lightweight Markdown-only skills.
β The AI Tool Bible editorial team
Pros
- β Lightweight Markdown-only skills
- β The vendor says that no framework or lock-in, working with various LLM agents
- β Supports cross-model review loops, idea discovery, and experiment automation
Facts
- Platforms
- cli, macos, windows (not source-checked)
- Pricing
- free - MIT license (not source-checked)
- API available
- Yes (not source-checked)
- Open source
- Yes (license: MIT) (not source-checked)
- GitHub stars
- 18,992 (checked Nov 11, 2026) (not source-checked)
- Last GitHub push
- Nov 8, 2026 (not source-checked)
- First commit
- Mar 2026 (not source-checked)
Where this tool sits
Use cases
Frequently asked
- How much does ARIS cost?
- ARIS is completely free and released under the MIT license. There are no subscription fees or hidden costs associated with using this methodology for your research workflows.
- Which development tools does ARIS integrate with?
- ARIS is compatible with Claude Code, Codex CLI, Cursor, Trae, Antigravity, GitHub Copilot CLI, OpenClaw, and DeepSeek Harness. It functions as a skill-based workflow within these environments.
- Does ARIS require specific API keys?
- No, ARIS supports alternative model combinations, including those hosted on ModelScope, which do not require Claude or OpenAI API keys. It uses an executor model and an independent review model.
- What are the main use cases for ARIS?
- ARIS is designed for autonomous ML research, cross-model review loops, idea discovery, experiment automation, paper review, and general research automation. It focuses on the research workflow itself.
- How is ARIS distributed?
- ARIS is available as a standalone CLI, as plugins, and through direct skill installations. It is a methodology rather than a standalone platform, focusing on lightweight Markdown-only skills.