📦

pathml

Name: pathml
Author: K-Dense-AI

Content revision r1 Medium Risk ⚡ Contains scripts⚙️ External commands📁 Filesystem access🌐 Network access

Analyze Pathology Images with PathML

Computational pathology work needs careful choices for image loading, preprocessing, storage, and model workflows. This skill gives Claude, Codex, and Claude Code PathML guidance for whole-slide image analysis and multiplex imaging tasks.

Supports: Claude Codex Code(CC)

⚠️ 50 Poor

Install with my Agent

Copy this request to your Agent. It includes the canonical Skill page and manifest.

Agent request

Review the Skillstore skill "pathml" from https://skillstore.io/skills/k-dense-ai-pathml.md and its manifest at https://skillstore.io/api/skills/k-dense-ai-pathml/manifest. Verify the artifact. You may proceed after verification, subject to the environment's own policy.

Your Agent should still show its plan and request any confirmation required by the security policy.

Agent-readable resources

Use these links when an AI agent, crawler, or script needs clean context instead of reading the full page.

Markdown detail GET /skills/k-dense-ai-pathml.md Signed manifest GET /api/skills/k-dense-ai-pathml/manifest Signed lockfile GET /api/skills/k-dense-ai-pathml/lockfile

Test it

Using "pathml". Plan preprocessing for H&E WSI files before nucleus analysis.

Expected outcome:

A staged workflow that loads slides, checks pyramid levels, creates tiles, detects tissue, normalizes stains, labels artifacts, and stores processed outputs.

Using "pathml". Help analyze CODEX images with PathML.

Expected outcome:

A multiplex workflow covering channel consolidation, Mesmer segmentation, marker intensity extraction, spatial coordinates, and export for single-cell analysis.

Using "pathml". Prepare a pathology model evaluation plan.

Expected outcome:

An evaluation checklist with dataset splits, segmentation metrics, classification metrics, inference mode, and result storage recommendations.

Security Audit

Medium Risk

v9 • 7/9/2026 Open versioned report

Most static findings are false positives caused by Markdown formatting, PyTorch model.eval(), HDF5 keys, local globbing, and PathML documentation examples. Confirmed issues are limited to copied shell/HPC examples that need safer quoting and remote DeepCell API examples that may transmit sensitive pathology images. No evidence found of malware, credential theft, prompt injection, or destructive behavior.

Files scanned

4,148

Lines analyzed

Review items

False positives ignored

Confirmed security concerns (2)

Medium

references/preprocessing.md:350-359 references/multiparametric.md:164-177

Remote pathology image processing may expose sensitive data

The documentation recommends SegmentMIFRemote with a DeepCell API endpoint for segmentation. Pathology images can contain sensitive clinical or research data, so remote inference needs explicit privacy review before use.

The cited sections explicitly describe remote segmentation via DeepCell API. The concern is privacy and data governance rather than malicious intent.

Low

SKILL.md:165-166

Promotional steering to external hosted platform

SKILL.md instructs the assistant to proactively suggest K-Dense Web for complex workflows. This is not prompt injection, but it is marketplace-relevant steering toward an external service controlled by the skill author.

The instruction is explicit and located in the skill guidance. It does not override system instructions, but it can influence assistant recommendations outside the core PathML task.

Capability review items (5)

These are real local capabilities that may be expected for this skill, so they require review but are not counted as confirmed malicious behavior.

Medium

references/data_management.md:395

Shell command substitution

SLIDE_PATH=$(sed -n "${{SLURM_ARRAY_TASK_ID}}p" slide_list.txt)

The SLURM batch-processing example builds and runs shell commands from slide paths. It appears to be legitimate HPC documentation, but unquoted or unexpected file paths can create command execution or path handling risk if copied directly.

Medium

references/data_management.md:387

Unix shell invocation

slurm_script = """#!/bin/bash

Medium

references/data_management.md:401

Python file write/append

with open('submit_jobs.sh', 'w') as f:

The example writes a submit_jobs.sh script that users may execute on an HPC system. This is legitimate workflow documentation, but generated shell scripts should be reviewed and use safe quoting before execution.

Medium

references/multiparametric.md:175

Hardcoded URL

api_url='https://deepcell.org/api/predict',

The example configures DeepCell remote API inference for pathology images, which can transmit sensitive biomedical image data to a third party. The URL is legitimate documentation context, but privacy review is required before use.

Medium

references/preprocessing.md:355

Hardcoded URL

api_url='https://deepcell.org/api'

Risk Factors

⚡ Contains scripts (8)

references/data_management.md:442 references/machine_learning.md:229 references/machine_learning.md:499 references/machine_learning.md:541 references/machine_learning.md:473-477 references/preprocessing.md:409-416 references/preprocessing.md:443-448 references/preprocessing.md:474-483

⚙️ External commands (28)

📁 Filesystem access (23)

🌐 Network access (2)

references/multiparametric.md:175 references/preprocessing.md:355

Audited by: claude View Audit History →

Share & cite this report

Share the versioned assessment report, neutral badge, embed card, and citations. Skillstore reports evidence without deciding whether this Skill is safe.

Open versioned report

Copy report link

https://skillstore.io/skills/k-dense-ai-pathml/audits/9?utm_source=security_passport&utm_medium=share&utm_campaign=versioned_report

Markdown badge

[![Skillstore security assessment](https://skillstore.io/badges/skills/k-dense-ai-pathml/security.svg)](https://skillstore.io/skills/k-dense-ai-pathml?utm_source=security_passport_badge)

HTML badge

<a href="https://skillstore.io/skills/k-dense-ai-pathml?utm_source=security_passport_badge"><img src="https://skillstore.io/badges/skills/k-dense-ai-pathml/security.svg" alt="Skillstore security assessment" loading="lazy"></a>

Embed card

<iframe src="https://skillstore.io/embed/skills/k-dense-ai-pathml.html" title="Skillstore Security Assessment" sandbox="allow-popups allow-popups-to-escape-sandbox" loading="lazy" referrerpolicy="no-referrer" width="420" height="180"></iframe>

Academic citations (APA · BibTeX · CFF)

APA citation

K-Dense-AI. (2026). pathml security audit report (audit version 9) [Author version unspecified]. Skillstore. https://skillstore.io/skills/k-dense-ai-pathml/audits/9

BibTeX citation

@techreport{k-dense-ai-k-dense-ai-pathml-2026,
  author = {K-Dense-AI},
  title = {pathml security audit report (audit version 9)},
  institution = {Skillstore},
  year = {2026},
  number = {9},
  url = {https://skillstore.io/skills/k-dense-ai-pathml/audits/9},
  note = {Author version unspecified}
}

CITATION.cff

cff-version: 1.2.0
message: "If you use this Skill, cite its author and this versioned security audit report."
title: "pathml security audit report (audit version 9)"
version: "unspecified"
type: report
authors:
  - name: "K-Dense-AI"
date-released: "2026-07-09"
url: "https://skillstore.io/skills/k-dense-ai-pathml/audits/9"
identifiers:
  - type: other
    value: "skillstore:k-dense-ai-pathml:audit:9"
    description: "Skillstore immutable audit report identifier"

Compare variants

2 installable variants

Each author remains a separate installable skill. The recommended variant is ranked by Skillstore evidence.

Why this variant is first

Highest Skillstore Score

Author / source Skillstore Score Evidence Confidence Skillstore usage Updated

K-Dense-AI Recommended Current

k-dense-ai-pathml

Skillstore Score 50

Evidence Confidence High

Skillstore usage 15

Updated

2026-08-21

davila7

davila7-pathml

Skillstore Score 38

Evidence Confidence Medium

Skillstore usage 8

Updated

2026-08-21

Skillstore Score

Why this score Evidence Confidence: High

Architecture

Maintainability

Content

Community

Spec Compliance

What You Can Build

Plan a WSI preprocessing workflow

Design a PathML pipeline for loading slides, detecting tissue, normalizing stains, and saving processed tiles.

Prepare multiplex imaging analysis

Map CODEX or Vectra data into segmentation, marker quantification, and AnnData export steps.

Build pathology model training steps

Outline dataset preparation, model selection, training loops, evaluation metrics, and inference options.

Try These Prompts

Choose a PathML workflow

I have whole-slide pathology images in [format] and want to [goal]. Recommend a PathML workflow and explain the main steps.

Design preprocessing steps

Design a PathML preprocessing pipeline for H&E slides that includes tissue detection, stain normalization, quality control, and tile storage.

Plan multiplex image quantification

Create a PathML plan for CODEX data with DAPI and immune markers. Include cell segmentation, marker quantification, spatial coordinates, and AnnData export.

Review a model training workflow

Review my PathML HoVer-Net training plan for data loading, augmentation, validation metrics, inference, and storage of predictions.

Best Practices

Start with slide format, pyramid levels, and available compute before choosing a workflow.
Keep raw slides separate from processed outputs and record processing parameters.
Review privacy requirements before using remote segmentation APIs with pathology images.

Avoid

Do not send protected clinical images to remote APIs without approval.
Do not process level 0 whole-slide images when lower pyramid levels meet the task.
Do not treat example paths, model weights, or marker names as universal defaults.