<2-3 sentences summarising what the project addresses and solves, the target audience, and the benefit.> Open Autonomous Generalist Scientist (OpenAGS) is an open-source framework designed for fully autonomous scientific research, covering the entire lifecycle from literature review to manuscript writing. It orchestrates a team of AI agents to accelerate scientific discovery, targeting researchers and AI engineers seeking to automate and enhance the scientific process. The project provides an end-to-end, multi-agent system for comprehensive research automation.

How It Works

OpenAGS employs a multi-agent architecture orchestrated by a Node.js server. This server integrates with various LLM APIs (Claude, Codex, Gemini) and external research tools like arXiv and Semantic Scholar. A React-based UI, accessible via Electron desktop app or browser, provides a workspace featuring chat, terminal, and a manuscript editor. The system handles workflow orchestration, experiment sandboxing using Docker, and communication with external services to manage the research process autonomously.

Quick Start & Requirements

• Primary install / run command: Clone the repository, cd OpenAGS, pnpm install. To launch the desktop app: cd packages/desktop && npx electron-vite dev. For server-only mode: pnpm --filter @openags/app dev.
• Non-default prerequisites and dependencies: Node.js (>= 20), pnpm (>= 9). Optional: TeX Live/BasicTeX (LaTeX compilation), Docker (sandboxed experiments), Rust (>= 1.75, for CLI development).
• Links: Documentation

Highlighted Details

• End-to-end autonomous research lifecycle support: literature review, hypothesis generation, experiments, manuscript writing, and peer review.
• Multi-agent collaboration framework for complex research tasks.
• Integrated LaTeX editor within the UI for manuscript preparation.
• Support for multiple LLM providers and research data sources.

Maintenance & Community

No specific details on notable contributors, sponsorships, or community channels (like Discord/Slack) are provided in the README.

Licensing & Compatibility

• License type: MIT License.
• Compatibility notes: The MIT license generally permits commercial use and linking with closed-source projects.

Limitations & Caveats

The Rust CLI component is noted as a "future" feature, indicating ongoing development. The README does not explicitly detail other limitations, alpha/beta status, or known bugs.