rdkit
Cheminformatics toolkit for fine-grained molecular control. SMILES/SDF parsing, descriptors (MW, LogP, TPSA), fingerprints, substructure search, 2D/3D generation, similarity, reactions. For standard workflows with simpler interface, use datamol (wrapper around RDKit). Use rdkit for advanced control, custom sanitization, specialized algorithms.
81
75%
Does it follow best practices?
Impact
93%
1.38xAverage score across 3 eval scenarios
Passed
No known issues
Optimize this skill with Tessl
npx tessl skill review --optimize ./scientific-skills/rdkit/SKILL.mdQuality
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 a strong skill description that excels across all dimensions. It lists specific concrete capabilities using domain-appropriate terminology that users would naturally use, clearly distinguishes itself from the related datamol skill, and provides explicit guidance on when to select this skill. The description is concise yet comprehensive, avoiding fluff while packing in actionable trigger terms.
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | Lists multiple specific concrete actions: SMILES/SDF parsing, descriptors (MW, LogP, TPSA), fingerprints, substructure search, 2D/3D generation, similarity, reactions. These are highly specific cheminformatics operations. | 3 / 3 |
Completeness | Clearly answers 'what' (cheminformatics toolkit with specific capabilities listed) and 'when' (use rdkit for advanced control, custom sanitization, specialized algorithms; distinguishes from datamol for simpler workflows). The 'Use rdkit for...' clause serves as explicit trigger guidance. | 3 / 3 |
Trigger Term Quality | Excellent coverage of natural terms a cheminformatics user would use: SMILES, SDF, molecular descriptors, MW, LogP, TPSA, fingerprints, substructure search, similarity, reactions, RDKit. These are the exact terms domain users would mention. | 3 / 3 |
Distinctiveness Conflict Risk | Highly distinctive niche in cheminformatics with clear differentiation from the related datamol skill. The description explicitly states when to use this skill vs. datamol, reducing conflict risk between the two most likely overlapping skills. | 3 / 3 |
Total | 12 / 12 Passed |
Implementation
50%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
This skill is highly actionable with excellent, executable code examples covering RDKit's full API surface. However, it is severely bloated—it reads more like comprehensive documentation than a concise skill file, with redundant explanations, duplicate patterns (Lipinski appears twice), and content that should be in reference files rather than inline. Workflow clarity suffers from missing validation checkpoints, particularly for operations that can silently fail (molecule parsing, embedding, reactions).
Suggestions
Reduce the main file to ~100-150 lines by moving detailed API examples (sections 3-12) into reference files, keeping only a quick-start overview and key patterns in SKILL.md
Remove all explanatory prose that Claude already knows (e.g., 'RDKit is a comprehensive cheminformatics library', 'Convert molecules to text representations', 'For processing multiple molecules, use Supplier/Writer objects')
Add explicit validation checkpoints to workflows: check EmbedMolecule return value (-1 = failure), verify reaction products are not None, validate sanitization success before proceeding
Deduplicate the Lipinski Rule of Five check which appears both in section 4 and in the Drug-likeness Analysis workflow
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | Extremely verbose at ~500+ lines. Contains extensive explanations Claude already knows (e.g., 'RDKit is a comprehensive cheminformatics library providing Python APIs for molecular analysis'), redundant code examples (Lipinski check appears twice), and explanatory text that adds no value for an AI agent ('Convert molecules to text representations'). Much of this could be cut by 60%+ while preserving all actionable content. | 1 / 3 |
Actionability | Provides fully executable, copy-paste ready Python code throughout. Code examples are concrete with real SMILES strings, specific function calls, proper imports, and complete workflow functions. The common workflows section provides end-to-end executable functions. | 3 / 3 |
Workflow Clarity | Steps within workflows are listed but lack explicit validation checkpoints. For example, the 3D coordinate generation workflow doesn't verify embedding success (EmbedMolecule returns -1 on failure), and reaction workflows don't validate products. The error handling section is separate rather than integrated into workflows. No feedback loops for error recovery. | 2 / 3 |
Progressive Disclosure | References to external files (api_reference.md, descriptors_reference.md, smarts_patterns.md, scripts/) are present at the end, but the main file is a monolithic wall of content that should have much more pushed to reference files. The 12 major sections with full code examples inline make this overwhelming when a concise overview with pointers would be more appropriate. | 2 / 3 |
Total | 8 / 12 Passed |
Validation
81%Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.
Validation — 9 / 11 Passed
Validation for skill structure
| Criteria | Description | Result |
|---|---|---|
skill_md_line_count | SKILL.md is long (779 lines); consider splitting into references/ and linking | Warning |
metadata_version | 'metadata.version' is missing | Warning |
Total | 9 / 11 Passed | |
- Repository
- K-Dense-AI/claude-scientific-skills
- Commit
086de41
- 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.