Devotd Github
NewAI agents

k-dense-ai/scientific-agent-skills v2.72.0 — A Collection of Science and Research Skills

k-dense-ai/scientific-agent-skills is a collection of skills for AI agents that help with science and research work.

Repositoryk-dense-ai/scientific-agent-skills ↗

k-dense-ai/scientific-agent-skills is a collection of skills for AI agents that help with science and research work.

GitHub description: Turn any AI agent into an AI Scientist. The #1 Agent Skills library for science, used by 250,000+ scientists worldwide. 177 ready-to-use validated skills plus 100+ scientific databases covering biology, chemistry, medicine, and drug discovery. Compatible with Cursor, Claude Code, Codex, Pi, Antigravity, and the open Agent Skills standard.

stars
★ 53.8K
Forks
4.7K
Language
Python
License
MIT
Latest release
v2.72.0
Read original source ↗View on GitHub ↗

The latest release recorded for k-dense-ai/scientific-agent-skills is v2.72.0. This repository is a collection of skills that help AI agents carry out science and research work. The skills provide usage guidelines and validation criteria for specific packages, data sources, and workflows. However, the provided material contains no changelog for v2.72.0, so what was newly added or modified in this version cannot be confirmed.

The README explains that it covers biology, chemistry, medicine, physics, engineering, earth sciences, data analysis, and science communication. The repository description states 177 ready-to-use validated skills and more than 100 scientific databases, including biology, chemistry, medicine, and drug development. It follows the Agent Skills standard, can be used in Cursor, Claude Code, Codex, Pi, Antigravity, and is also offered as an Agent Plugins package. It can therefore be viewed as a reference that connects specific research tools and information sources to agent workflows.

That said, no validation criteria or actual performance data for the 177 skills are given. The figure of more than 250,000 scientists worldwide using it is also a claim stated in the repository description, with no separate supporting evidence. With no release changelog either, it is impossible to tell whether v2.72.0 expanded functionality or corrected existing guidance. The range of possible uses is broad, but the accuracy of the skills and the degree of improvement in this version are difficult to evaluate from this material alone.