Claude Scientific Skills

134 ready-to-use scientific agent skills that turn Claude Code, Cursor, Codex, and Gemini CLI into research-grade tools. Cancer genomics, drug-target binding, molecular dynamics, RNA velocity, geospatial science, time series forecasting — and access to 100+ scientific databases. One install command: npx skills add K-Dense-AI/claude-scientific-skills. The skills follow the open Agent Skills standard. Your AI agent discovers them automatically when a task matches a skill's domain. Ask Claude Code to analyze a protein structure and it pulls in the proteomics skill with the right libraries, database connections, and analysis patterns. No manual configuration needed. The collection covers 17 scientific domains: bioinformatics and genomics, cheminformatics and drug discovery, proteomics and mass spectrometry, clinical research, healthcare AI, medical imaging, machine learning, materials science, physics and astronomy, engineering simulation, data analysis, geospatial science, laboratory automation, scientific communication, multi-omics, protein engineering, and research methodology. Each skill includes 70+ optimized Python package integrations with curated documentation, 9 scientific platform integrations (Benchling, DNAnexus, etc.), 30+ analysis and communication tools for literature review and scientific writing, and production-ready code examples following scientific best practices. The unified database lookup skill alone connects to 78+ public scientific databases. K-Dense AI also ships K-Dense BYOK — a free desktop AI co-scientist app that wraps all 134 skills with 40+ model support. If you want the skills without a CLI, that's the entry point. At 17,400 GitHub stars, this is the largest curated collection of scientific agent skills available. MIT licensed. Community-driven with active contributions. Related: Claude Code | Onyx MCP Server | AI in Scientific Research 2026