datamol

Pythonic wrapper around RDKit with simplified interface and sensible defaults. Preferred for standard drug discovery including SMILES parsing, standardization, descriptors, fingerprints, clustering, 3D conformers, parallel processing. Returns native rdkit.Chem.Mol objects. For advanced control or custom parameters, use rdkit directly.

3.24x

Quality

66%

Does it follow best practices?

Impact

81%

3.24x

Average score across 3 eval scenarios

Securityby

Advisory

Suggest reviewing before use

Optimize this skill with Tessl

npx tessl skill review --optimize ./scientific-skills/datamol/SKILL.md

Evaluation results

97%

73%

Virtual Screening: Find Compounds Similar to a Known Active

Molecular similarity search and diversity selection

Criteria

Without context

With context

Datamol import alias

100%

Molecule standardization

100%

Standardization parameters

100%

None check after parsing

100%

ECFP fingerprint type

62%

Cross-set distance with cdist

100%

Parallel processing

100%

Diversity selection

100%

Visualization output

100%

Visualization legends

100%

hits.csv output

100%

diverse_hits.png output

100%

55%

28%

Preparing a Molecular Dataset for Machine Learning

Descriptor computation, drug-likeness filtering, scaffold-based ML splitting

Criteria

Without context

With context

Datamol import alias

100%

Standardization with all params

100%

Batch descriptor computation

100%

Parallel descriptor computation

Lipinski MW threshold

100%

Lipinski LogP threshold

Lipinski HBD threshold

Lipinski HBA threshold

Murcko scaffold extraction

100%

Scaffold-based split

100%

Output files present

100%

Split report content

100%

93%

69%

3D Conformer Analysis for Lead Optimization

3D conformer generation, clustering, and SASA analysis

Criteria

Without context

With context

Datamol import alias

100%

ETKDGv3 method

100%

Energy minimization

100%

Conformer clustering

100%

SASA computation

100%

SASA property access

100%

Parallel SASA

Conformer visualization

100%

conformer_report.json output

100%

Report required fields

100%

conformer_grid.png output

100%

Repository: K-Dense-AI/claude-scientific-skills
Commit: b58ad7e

Evaluated: 27 days ago
Agent: Claude Code
Model: Claude Sonnet 4.6

Table of Contents

Virtual Screening: Find Compounds Similar to a Known Active Preparing a Molecular Dataset for Machine Learning 3D Conformer Analysis for Lead Optimization

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.