Collection of scientific skills for working with specialized libraries, databases, and research workflows.
What it does
Scientific computing has specialised library APIs that Claude approximates rather than knows — BioPython sequences have specific indexing, RDKit molecules require a specific sanitisation workflow before operations, and Astropy units don't just behave like Python numbers. This skill loads the correct APIs and workflow patterns for specialised scientific libraries and research databases: genomics, chemistry, astronomy, physics simulation, and the specific patterns for accessing scientific databases (PubMed, ChEMBL, NASA ADS) programmatically. Made by K-Dense-AI.
Use case
Research workflows that require specialised scientific libraries or programmatic access to scientific databases. Most valuable when Claude keeps generating code that looks correct but fails when you actually run it against real scientific data.
"Parse this FASTA file and compute the GC content of each sequence." "Load this molecule with RDKit and compute its fingerprint for similarity search." "Query ChEMBL for all approved drugs that target this protein." "Calculate the orbital parameters for this object using its ephemeris data." "Run a molecular dynamics simulation setup for this protein structure."
Describe the scientific task and which libraries or databases you're working with.
Claude generates code using correct API patterns for the specific library version.
For database queries: describe what data you need and Claude generates the correct query structure.
Input
A scientific computing task, the relevant data format or file, and which libraries or databases are involved.
Output
Scientific computing code using correct library APIs — proper BioPython sequence handling, RDKit with sanitisation, Astropy with units, and correct database query patterns that return real results.
npx skillsadd K-Dense-AI/skills/scientific-research
Requires skills.sh CLI
Comprehensive LLM training skill with guidance, helper scripts, cost estimators, and best practices.
Orchestrate ML model evaluation jobs — benchmarks, metrics, reports, and comparison dashboards.
Techniques for context engineering with AI agents — managing context windows, prompt structure, and knowledge organisation.