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.

Quality

80%

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

68%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 is highly actionable with comprehensive executable examples and a well-organized reference/script bundle, but it is verbose due to inlined API-reference material and lacks explicit validation feedback loops for its batch workflows.

Suggestions

Move the per-capability API tours (e.g., full fingerprint-generator options, drawing customization, pharmacophore factory) into the existing references and keep SKILL.md as a lean overview with key examples, improving conciseness and progressive_disclosure.

Add explicit validation checkpoints and feedback loops to the batch workflows — e.g., in similarity_screen and the SDF suppliers, report/log skipped None molecules and verify output integrity before proceeding — to lift workflow_clarity above the batch-cap of 3.

Replace deprecated API usage (e.g., AllChem.GetMorganFingerprintAsBitVect) consistently with the modern rdFingerprintGenerator API already shown elsewhere in the skill to keep examples current.

Dimension	Reasoning	Score
Conciseness	The ~770-line body is dense and code-driven without concept padding, but substantial API-reference-style material (fingerprint generators, drawing customization, pharmacophore factory) is inlined when equivalent detail lives in the references files, so it could be tightened.	3 / 5
Actionability	Provides extensive copy-paste-ready, executable code with real function calls and parameters across all capabilities, plus complete workflow functions (analyze_druglikeness, similarity_screen, filter_by_substructure) covering the common cases.	5 / 5
Workflow Clarity	Workflow functions are sequenced with basic None-guards, but batch/destructive operations (SDF batch processing, database similarity screening, list filtering) lack explicit validation checkpoints or validate→fix→retry feedback loops, capping this dimension at 3 per the rubric.	3 / 5
Progressive Disclosure	Three real references and three real scripts are clearly listed and signaled in a Resources section with one-level-deep navigation, but significant API-reference content remains inlined in the body rather than being fully split out.	4 / 5
	Total	15 / 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.

The description is specific, third-person, and answers both what the skill does and when to use it, with explicit differentiation from the related datamol skill. Trigger term coverage is strong but could include a few more natural synonyms.

Dimension	Reasoning	Score
Specificity	Enumerates multiple concrete actions — 'SMILES/SDF parsing, descriptors (MW, LogP, TPSA), fingerprints, substructure search, 2D/3D generation, similarity, reactions' — giving comprehensive coverage of the toolkit's capabilities.	5 / 5
Completeness	Explicitly states what it does ('Cheminformatics toolkit for fine-grained molecular control…') and when to use it ('Use rdkit for advanced control, custom sanitization, specialized algorithms'), with concrete trigger phrases.	5 / 5
Trigger Term Quality	Strong natural keyword coverage (SMILES, SDF, descriptors, fingerprints, substructure search, similarity) including file extensions, but misses a few common synonyms users say (e.g., ECFP/Morgan, .sdf).	4 / 5
Distinctiveness Conflict Risk	Carves a clear niche and explicitly distinguishes from a sibling skill ('For standard workflows with simpler interface, use datamol… Use rdkit for advanced control'), minimizing conflict risk.	5 / 5
	Total	19 / 20 Passed

Validation

81%

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 — 13 / 16 Passed

Validation for skill structure

Criteria	Description	Result
skill_md_line_count	SKILL.md is long (784 lines); consider splitting into references/ and linking	Warning
metadata_version	'metadata.version' is missing	Warning
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	13 / 16 Passed

Repository: synthetic-sciences/openscience
Path: backend/cli/skills/chemistry/rdkit/SKILL.md
Commit: 7dabbf0

Reviewed: 2 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.