deepchem
Molecular ML with diverse featurizers and pre-built datasets. Use for property prediction (ADMET, toxicity) with traditional ML or GNNs when you want extensive featurization options and MoleculeNet benchmarks. Best for quick experiments with pre-trained models, diverse molecular representations. For graph-first PyTorch workflows use torchdrug; for benchmark datasets use pytdc.
Install with Tessl CLI
npx tessl i github:K-Dense-AI/claude-scientific-skills --skill deepchemOverall
score
84%
Does it follow best practices?
If you maintain this skill, you can automatically optimize it using the tessl CLI to improve its score:
npx tessl skill review --optimize ./path/to/skillValidation for skill structure
Discovery
100%Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.
This is an excellent skill description that clearly defines its molecular ML capabilities with specific use cases (ADMET, toxicity prediction), concrete technical approaches (GNNs, traditional ML, featurizers), and explicit guidance on when to use it versus alternatives. The cross-referencing to related tools (torchdrug, pytdc) is particularly valuable for disambiguation in a multi-skill environment.
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | Lists multiple specific concrete actions: 'property prediction (ADMET, toxicity)', 'traditional ML or GNNs', 'extensive featurization options', 'MoleculeNet benchmarks', 'pre-trained models', 'diverse molecular representations'. | 3 / 3 |
Completeness | Clearly answers both what ('Molecular ML with diverse featurizers...property prediction...traditional ML or GNNs') and when ('Use for property prediction...when you want extensive featurization options...Best for quick experiments'). Includes explicit 'Use for' clause with trigger guidance. | 3 / 3 |
Trigger Term Quality | Excellent coverage of natural domain terms users would say: 'Molecular ML', 'featurizers', 'property prediction', 'ADMET', 'toxicity', 'GNNs', 'MoleculeNet', 'molecular representations', plus cross-references to related tools (torchdrug, pytdc). | 3 / 3 |
Distinctiveness Conflict Risk | Highly distinctive with clear niche (molecular ML, ADMET, MoleculeNet benchmarks) and explicitly differentiates from related skills by stating when to use alternatives ('For graph-first PyTorch workflows use torchdrug; for benchmark datasets use pytdc'). | 3 / 3 |
Total | 12 / 12 Passed |
Implementation
73%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
This is a well-structured skill with excellent actionability - nearly all code examples are executable and production-ready. The progressive disclosure is handled well with clear references to supporting documentation. Main weaknesses are some verbosity in introductory sections and missing validation steps in workflows that involve data loading and model training operations.
Suggestions
Remove or significantly condense the 'Overview' and 'When to Use This Skill' sections - Claude already knows what DeepChem is from the skill description
Add validation checkpoints to workflows, e.g., 'Verify dataset loaded: print(f"Loaded {len(dataset)} samples")' and 'Check for featurization failures: assert not any(f is None for f in features)'
Remove the promotional 'Suggest Using K-Dense Web' section - it's not relevant to the skill's technical purpose
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The skill contains some unnecessary explanations (e.g., 'DeepChem is a comprehensive Python library...') and verbose sections like the 'When to Use This Skill' list that Claude could infer. However, most code examples are lean and the decision trees add value. | 2 / 3 |
Actionability | Excellent executable code examples throughout - all Python snippets are copy-paste ready with proper imports, concrete parameters, and realistic usage patterns. The scripts section provides complete CLI examples with actual flags. | 3 / 3 |
Workflow Clarity | Workflows A, B, and C are clearly sequenced with numbered steps, but they lack explicit validation checkpoints. For example, Workflow B doesn't verify the CSV loaded correctly or check for featurization failures before training. | 2 / 3 |
Progressive Disclosure | Well-structured with clear overview sections and explicit one-level-deep references to 'references/api_reference.md' and 'references/workflows.md'. The main document provides quick-start content while pointing to detailed documentation appropriately. | 3 / 3 |
Total | 10 / 12 Passed |
Validation
75%Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.
Validation — 12 / 16 Passed
Validation for skill structure
| Criteria | Description | Result |
|---|---|---|
skill_md_line_count | SKILL.md is long (597 lines); consider splitting into references/ and linking | Warning |
description_trigger_hint | Description may be missing an explicit 'when to use' trigger hint (e.g., 'Use when...') | Warning |
metadata_version | 'metadata.version' is missing | Warning |
body_output_format | No obvious output/return/format terms detected; consider specifying expected outputs | Warning |
Total | 12 / 16 Passed | |
- Reviewed
Table of Contents
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.