Datamol

This is a Pythonic wrapper around RDKit that handles the everyday cheminformatics grunt work: parsing SMILES, standardizing structures, computing descriptors and fingerprints, clustering compounds, and generating 3D conformers. It returns native RDKit Mol objects so you can drop down to raw RDKit when you need fine control. The API is cleaner than vanilla RDKit (compare `dm.to_mol()` and `dm.standardize_mol()` to the RDKit equivalents), includes built-in parallelization for batch operations, and supports reading from S3 or GCS via fsspec. Good for drug discovery workflows where you're processing datasets and don't want to write boilerplate. Requires Python 3.8+ and installs RDKit automatically as a dependency.

Install to Claude Code

npx -y skills add k-dense-ai/scientific-agent-skills --skill datamol --agent claude-code