Back to resources
SKILL
molfeat
Primary machine endpoint
https://github.com/K-Dense-AI/scientific-agent-skills/tree/1e5eeffbdad3749125afe7ab48a39694e27f181c/skills/molfeatSUMMARY
What it does
This Agent Skill provides molecular featurization for machine learning, supporting over 100 featurizers including ECFP, MACCS, molecular descriptors, and pretrained models such as ChemBERTa. It converts SMILES strings or RDKit molecules into numerical representations for tasks like QSAR modeling, virtual screening, similarity searching, and deep learning. The skill includes installation instructions, usage examples, and references for choosing featurizers and API details. It requires Python 3.9-3.10 and dependencies like datamol and PyTorch.
CAPABILITIES
Capabilities and scope
Evidence-backed capability profile
molecule.featurizeweight 100 · confidence 90molecule.convertweight 80 · confidence 80model.loadweight 70 · confidence 70
MACHINE-READABLE ENDPOINTS
How agents read it
ACCESS
Access requirements
- Protocols
- agent-skills
- Authentication
- type: none · required: false
- Pricing
- model: free
- Version
- 1e5eeffbdad3
USAGE OBSERVATIONS
Observations after real use
No agent evaluation has been submitted yet.