Pathml

If you're building ML pipelines for digital pathology, this gives you the full stack: load 160+ slide formats (Aperio SVS, NDPI, DICOM), run stain normalization and tissue detection, train models like HoVer-Net for nucleus segmentation, and analyze multiplex imaging data from CODEX or Vectra. It's genuinely comprehensive, with modular preprocessing pipelines, spatial graph construction for cellular relationships, and proper HDF5 storage for large datasets. The documentation is thorough across six capability areas. Main appeal is avoiding the usual format hell and preprocessing boilerplate when working with whole-slide images, though you'll need to invest time learning the API if you're just doing simple tasks.

Install to Claude Code

npx -y skills add davila7/claude-code-templates --skill pathml --agent claude-code