torchdrug-english

PyTorch-native Graph Neural Network framework for molecules and proteins. Suitable for building custom GNN architectures for drug discovery, protein modeling, or knowledge graph reasoning. Best for custom model development, protein property prediction, and retrosynthesis. If you need pretrained models and diverse feature extractors, use deepchem; if you need benchmark datasets, use pytdc.

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/TorchDrug-English/SKILL.md

Quality

Content

53%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 body pairs an executable quick-start with well-organized one-level-deep references, but is weighed down by duplicated description text, generic boilerplate, and missing validation steps in the workflows.

Suggestions

Remove the verbatim description duplicated in the "When to Use" and "Key Features" sections and drop the generic "Use this skill when a data analytics task needs a packaged method" boilerplate, which applies to any skill rather than TorchDrug.

Add an explicit validation/checkpoint step to each Common Workflow (e.g., verify train/val metrics before declaring a model trained, confirm scaffold split ran) so workflow_clarity can exceed 3.

Convert the "Reference: See X.md" pointers into markdown links and add a pointer to the existing references/core_concepts.md so all bundle files are reachable and navigation matches the 5-anchor.

Dimension	Reasoning	Score
Conciseness	Noticeably verbose: the description is duplicated verbatim in both the "When to Use" and "Key Features" sections, generic boilerplate bullets ("Use this skill when a data analytics task needs a packaged method...") apply to any skill, and "Implementation Details" offers platitudes ("keep results reproducible, identify assumptions explicitly").	2 / 5
Actionability	A complete, copy-paste-ready Quick Example plus concrete workflows with real API calls (datasets.BBBP(), models.GIN, tasks.PropertyPrediction) give mostly executable guidance with only minor gaps.	4 / 5
Workflow Clarity	Workflows list a clear sequence (load dataset -> choose model -> define task -> train -> evaluate) but contain no validation checkpoints or feedback loops, matching the 3-anchor's "checkpoints missing or implicit."	3 / 5
Progressive Disclosure	References are one level deep and clearly signaled per capability ("Reference: See molecular_property_prediction.md"), with navigation lines into reference sub-sections; minor gaps are the unreferenced core_concepts.md and plain-text refs instead of markdown links.	4 / 5
	Total	13 / 20 Passed

Description

78%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.

A strong, third-person description that conveys a clear niche and even routes users to sibling skills. It would reach the top level by adopting an explicit "Use when..." user-mention trigger clause.

Dimension	Reasoning	Score
Specificity	Names the domain and several concrete capabilities ("building custom GNN architectures for drug discovery, protein modeling, or knowledge graph reasoning", "protein property prediction, and retrosynthesis"), but they are framed more as use cases than the crisp action verbs of the 5-anchor.	4 / 5
Completeness	Clearly answers what ("PyTorch-native Graph Neural Network framework for molecules and proteins") and when ("Suitable for...", "Best for...", "If you need... use deepchem/pytdc"), but the when lacks the literal "Use when the user mentions X" trigger phrasing of the 5-anchor.	4 / 5
Trigger Term Quality	Good coverage of natural domain terms ("drug discovery", "retrosynthesis", "protein modeling") plus explicit conditional routing ("If you need pretrained models... use deepchem"), though it lacks file extensions and common synonym variants.	4 / 5
Distinctiveness Conflict Risk	Clear niche (PyTorch-native GNN for molecules/proteins) with explicit disambiguation against deepchem and pytdc, giving minimal conflict risk.	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: 2 missing	Warning

	Total	14 / 16 Passed

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

Reviewed: 4 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.