pathml

A full-featured computational pathology toolkit for advanced WSI analysis, including multiplexed immunofluorescence (CODEX, Vectra), nuclei segmentation, tissue graph construction, and machine learning model training on pathology data. Supports over 160 slide formats. For simp...

Quality

66%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Passed

No findings from the security scan

Fix and improve this skill with Tessl

tessl review fix ./scientific-skills/Data Analysis/pathml/SKILL.md

Quality

Content

57%Weight 40%Scale 1-5

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

The content has a solid executable Quick Start and clean capability-based structure, but it is padded with duplicated template sections and critically depends on references/*.md files that are not present in the bundle. Removing boilerplate, shipping the referenced files, and adding concrete validation checkpoints would materially improve it.

Suggestions

Either include the referenced reference files (references/image_loading.md, preprocessing.md, graphs.md, machine_learning.md, multiparametric.md, data_management.md) or remove the broken pointers — currently every 'See references/...' link dead-ends.

Delete the duplicated/generic sections (the second 'When to Use This Skill', 'Overview' restating the description, 'Implementation Details' pointing back to Overview, and the promotional 'K-Dense Web' block) to tighten token efficiency.

Add explicit validation feedback loops for the batch/destructive workflows — e.g. a concrete 'verify slide loaded / check tile shape / re-run on failure' checkpoint in the preprocessing and model-training workflows.

Dimension	Reasoning	Score
Conciseness	Mostly efficient with useful Quick Start code, but padded with generic template boilerplate and duplication — two 'When to Use' sections, an 'Overview' that restates the description, and 'Implementation Details' that simply points back to '## Overview above'.	3 / 5
Actionability	The Quick Start provides executable, copy-paste-ready bash and Python (SlideData, Pipeline, transforms) covering the common case, but the many 'See references/x.md' pointers are broken (no references/ directory exists), leaving gaps.	4 / 5
Workflow Clarity	Sequenced steps exist (Common Workflows, run plan), but for batch/destructive operations (WSI processing, ML training) there are no concrete validate->fix->retry checkpoints with commands — only generic prose — capping this at 3.	3 / 5
Progressive Disclosure	The body is well structured with clearly signaled one-level-deep references per capability area, but every referenced file (references/image_loading.md, preprocessing.md, etc.) is missing from the bundle, so navigation dead-ends.	3 / 5
	Total	13 / 20 Passed

Description

75%Weight 40%Scale 1-5

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

The description is specific and well-targeted to a distinct niche, but is literally truncated and lacks any explicit 'when to use' trigger guidance, which caps completeness. Adding a concrete 'Use when...' clause and completing the cut-off text would raise the score.

Suggestions

Complete the truncated description (it ends at 'For simp...') and add an explicit 'Use when...' trigger clause naming concrete user scenarios (e.g. loading whole slide images, training HoVer-Net, CODEX marker quantification).

Include a few natural synonyms and file extensions users say ('whole slide images', '.svs', '.ndpi', 'H&E staining') to broaden trigger-term coverage.

Keep the third-person voice while tightening the capability list so the description stays scannable.

Dimension	Reasoning	Score
Specificity	Names the computational pathology domain and lists multiple concrete actions — 'multiplexed immunofluorescence (CODEX, Vectra), nuclei segmentation, tissue graph construction, and machine learning model training' — for comprehensive coverage. Third-person voice is used correctly.	5 / 5
Completeness	Clearly answers 'what' with concrete capabilities, but there is no 'Use when...' clause or equivalent explicit trigger guidance (the field ends mid-sentence at 'For simp...'), capping completeness at 3 per the guidelines.	3 / 5
Trigger Term Quality	Good keyword coverage with domain-natural terms ('WSI analysis', 'nuclei segmentation', 'CODEX', 'Vectra', 'pathology data'), but missing common synonyms/extensions (e.g. 'whole slide image' spelled out, .svs/.ndpi file formats).	4 / 5
Distinctiveness Conflict Risk	Clear, specialized niche (computational pathology / WSI / multiplexed IF) with distinct triggers and minimal overlap risk; it even begins to differentiate from histolab.	5 / 5
	Total	17 / 20 Passed

Validation

87%

Warnings & errors only

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 14 / 16 Passed

Validation for skill structure

Criteria	Description	Result
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning
referenced_paths_exist	Referenced path issues: 14 missing	Warning

	Total	14 / 16 Passed

Repository: aipoch/medical-research-skills
Path: scientific-skills/Data Analysis/pathml/SKILL.md
Commit: 63c61d3

Reviewed: 10 hours ago

Table of Contents

Discovery Implementation Validation

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.