torchdrug

Build and troubleshoot TorchDrug 0.2.1 workflows for molecular graphs, property prediction, self-supervised pretraining, molecule generation, retrosynthesis, protein representation learning, and knowledge graph reasoning. Use when code imports torchdrug or needs its datasets, models, tasks, or Engine.

Quality

89%

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

Quality

Content

86%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.

A well-structured, actionable skill body with executable canonical code, a clear four-step workflow, a version guard, and clean one-level reference disclosure to real bundle files. Slight verbosity and the absence of an explicit in-workflow validation loop are the only drags.

Dimension	Reasoning	Score
Conciseness	Mostly efficient and assumes Claude's competence (no basic-concept padding), but the upstream-sources and citing sections plus the breadth of workflow listings add length that could be trimmed slightly; not a fully lean 5.	4 / 5
Actionability	Provides copy-paste ready installation commands and a complete executable property-prediction workflow, with specific API names (GIN, PropertyPrediction, GCPNGeneration, RotatE, etc.) and concrete gotchas covering the common cases.	5 / 5
Workflow Clarity	The 4-step load→model→task→Engine sequence is stated up front with a version-guard checkpoint and a 7-rule checklist, but the main training workflow lacks an explicit validate-then-proceed feedback loop, keeping it below 5.	4 / 5
Progressive Disclosure	SKILL.md is a clear overview with well-signaled one-level-deep references ('Read [molecular property prediction](references/molecular_property_prediction.md)'), and all 8 referenced files exist in ./references/, so navigation is easy and appropriately split.	5 / 5
	Total	18 / 20 Passed

Description

92%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 clearly states capabilities across multiple TorchDrug workflow domains and gives an explicit 'Use when' trigger tied to importing torchdrug. Minor room for more synonymous trigger phrasing, but it is comprehensive and distinct.

Dimension	Reasoning	Score
Specificity	Lists multiple concrete workflow domains (property prediction, self-supervised pretraining, molecule generation, retrosynthesis, protein representation learning, knowledge graph reasoning) plus build/troubleshoot actions, giving comprehensive coverage rather than 1-2 actions.	5 / 5
Completeness	Explicitly states both what (build/troubleshoot TorchDrug 0.2.1 workflows for the listed domains) and when ('Use when code imports torchdrug or needs its datasets, models, tasks, or Engine') with concrete trigger phrases.	5 / 5
Trigger Term Quality	Includes natural triggers like 'imports torchdrug' and 'datasets, models, tasks, or Engine' that users would say, but lacks synonyms or extension-style variants that would push it to comprehensive coverage.	4 / 5
Distinctiveness Conflict Risk	Targets a clear niche (TorchDrug 0.2.1 specifically) with a distinct import-based trigger, so overlap with generic PyTorch or cheminformatics skills is minimal.	5 / 5
	Total	19 / 20 Passed

Validation

100%

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 — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository: K-Dense-AI/scientific-agent-skills
Path: skills/torchdrug/SKILL.md
Commit: 1e5eeff

Reviewed: 3 days 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.