K-Dense-AI/scientific-agent-skills
163 how-to manuals that turn AI coding assistants into science research helpers
This repository packages 163 'skills' — structured instructions plus example code — that teach AI coding agents like Cursor, Claude Code, and Codex how to correctly use specific scientific software and databases. It covers over 100 databases such as PubChem, UniProt, and ClinicalTrials.gov, and more than 70 Python packages like RDKit and Scanpy. Every skill ships with documentation, examples, and its own test suite that runs in continuous integration.
What it does
- Provides 163 SKILL.md instruction sets spanning drug discovery, single-cell RNA analysis, variant annotation, clinical trial search, and more across biology, chemistry, medicine, and physics
- Each skill bundles documentation, code examples, and tests; any skill with scripts must have a matching test suite, and pull requests without one are blocked
- Installable via npx, GitHub CLI (gh skill), or manual copy into an agent's skills folder, working across Cursor, Claude Code, Codex, and other Agent Skills-compatible hosts
- Includes the docx/pdf/pptx/xlsx document skills originally built by Anthropic, and runs weekly security scans via the Cisco AI Defense Skill Scanner
- Warns that installed skills can make the AI agent run arbitrary code, install packages, and reach the network, so users should read each SKILL.md before installing
Why it matters
Instead of researchers re-learning how to call each scientific library or database from scratch, they can hand an AI agent a pre-vetted playbook for that task. But because skills can trigger code execution and network access, reviewing them before install matters for safety.
Terms in this repo
- Agent Skills · an open standard for packaging step-by-step instructions that AI agents can load and follow
- SKILL.md · the documentation file inside each skill describing what it does and how to use it
- database-lookup · a single skill that gives unified access to 78+ public scientific databases
- uv · the package manager used to install a skill's Python dependencies
Repository description (English)
Turn any AI agent into an AI Scientist. The #1 Agent Skills library for science, used by 175,000+ scientists worldwide. 163 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.
Open on GitHubTrending repos
- tt-a1i/archifyAn AI agent skill that turns a described codebase into a verified, shareable architecture diagram inside chat
- bookorbit/bookorbitA self-hosted book server that syncs your reading progress across web, Kobo, and KOReader
- ConardLi/garden-skillsA skill pack that hands AI coding agents ready-made expertise for design, video, images, and search
- browser-use/browser-useA tool that lets AI click buttons and type into websites just like a human, to finish tasks for you
- DietrichGebert/ponytailA ruleset that teaches AI coding agents to say no to writing code
- liustack/modlensA plugin that lets text-only AI models like DeepSeek 'see' pasted images by turning them into structured text data
- marin-community/marinMarin is an open-source platform that publishes the entire process of building large language models, not just the final model
- TauricResearch/TradingAgentsA framework where multiple LLM agents split into analyst, researcher, trader and risk-management roles, debate, and decide stock trades
Latest from METAL LAB
- Ox Alpha turns out to be GLM-5.3-Flash
- Martin Casado says AI has erased startups' distribution barrier
- OpenAI introduces Runme, which uses Codex to document repetitive work
- Stable Diffusion Maker Raises $76M From Music, Game Studios
- Meta's paid agent Hatch to launch alongside new Watermelon model in October