Best Research & science harnesses for AI agents
The most-adopted Research & science harnesses an AI agent can use, ranked by GitHub stars, with what each is best for. Loadbay is an MCP server, so an agent can pull this list live:
claude mcp add --transport http loadbay https://loadbay.xyz/api/mcp
-
1. MiroFish
72,897★ ·
Most adopted — the default starting point. Best for OASIS, CAMEL-AI, OpenAI-compatible APIs. Multi-agent swarm simulation engine that builds a parallel digital world from real-world seed signals, lets thousands of agents with memory interact, and predicts outcomes. Built on OASIS/CAMEL-style social simulation for decision rehearsal. -
2. scientific-agent-skills
33,496★ ·
Best for RDKit, Scanpy, PyTorch Lightning. A library of 161 pre-built agent skills covering biology, chemistry, genomics, drug discovery, and clinical research, compatible with Claude Code, Cursor, and Codex. Agents install skills via npx to gain access to 100+ scientific databases and specialized research analysis workflows. -
3. DeepTutor
28,764★ ·
Best for Claude Code, OpenAI Codex, OpenAI. Agent-native learning workspace explicitly designed to be operated by other AI agents, connecting tutoring, research, quiz generation, visualization, and mastery practice with JSON output modes for agent integration. -
4. gpt-researcher
28,053★ ·
Best for OpenAI, Anthropic, Tavily Search. Autonomous deep-research agent that plans and executes multi-step web research on any topic, synthesizing comprehensive reports from multiple LLMs and sources that other agents can act on. -
5. AI-Scientist
14,000★ · Python
Best for OpenAI, Anthropic, Semantic Scholar. First end-to-end system for fully automated open-ended scientific discovery: generates ideas, runs experiments, and writes papers. -
6. STORM
14,000★ · Python
Best for OpenAI, Bing, You.com. Stanford knowledge-curation system that researches a topic via internet search and writes a full-length, cited report. -
7. Feynman
9,183★ ·
Best for Codex, LM Studio, Ollama. Open-source AI research agent CLI with a workbench, local and hosted model providers, and installable research skills for Codex. -
8. paper-qa
8,722★ · Python
Best for arXiv, Semantic Scholar. High-accuracy RAG framework for answering questions from scientific papers with grounded citations. -
9. AlphaFold 3
8,200★ · Python
Best for JAX, PDB. Official DeepMind inference pipeline for AlphaFold 3, predicting joint structures of proteins, nucleic acids, ligands, and ions. -
10. AI-Scientist-v2
6,600★ · Python
Best for OpenAI, Anthropic, Semantic Scholar. Agentic successor that uses progressive tree search to autonomously generate hypotheses, run ML experiments, and write manuscripts.