Back to resources

SKILL

pathml

Primary machine endpointhttps://github.com/K-Dense-AI/scientific-agent-skills/tree/1e5eeffbdad3749125afe7ab48a39694e27f181c/skills/pathml
Use with an agent

SUMMARY

What it does

This skill provides guidance for using PathML in local computational pathology research. It covers loading and tiling whole-slide images, building preprocessing and QC pipelines, managing h5path data, quantifying multiplex images, constructing spatial graphs, and planning bounded model inference. The skill emphasizes safety boundaries, including de-identification, no-network default, and model-code security. It includes installation instructions for PathML 3.0.5, a stable minimal workflow example, and references to detailed documentation. The skill is MIT-licensed and authored by K-Dense Inc.

CAPABILITIES

Capabilities and scope

Evidence-backed capability profile

skill.pathml.load-and-tile-slidesweight 90 · confidence 90skill.pathml.build-preprocessing-pipelinesweight 90 · confidence 90skill.pathml.manage-h5path-dataweight 80 · confidence 80skill.pathml.quantify-multiplex-imagesweight 70 · confidence 70skill.pathml.construct-spatial-graphsweight 70 · confidence 70skill.pathml.plan-bounded-model-inferenceweight 80 · confidence 80

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.