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

86%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Low

Low-risk findings worth noting

Quality

Content

72%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 well-structured and efficiently offloads detail to verified bundle files, but it lacks a sequenced workflow with explicit validation checkpoints and leans on references rather than providing inline worked examples.

Suggestions

Add a short numbered workflow for a representative multi-step task (e.g. parse → validate with DetectChemistryProblems → sanitize → compute descriptors) with explicit validate-then-proceed checkpoints to raise workflow_clarity above 3.

Include one or two compact inline code snippets for the most common operations so the body is actionable without requiring a reference load.

Move the time-sensitive RDKit version baseline out of the narrative overview and into a clearly labeled compatibility/versioning note so the overview stays evergreen.

Dimension	Reasoning	Score
Conciseness	The body is dense and avoids explaining concepts Claude already knows, but inline time-sensitive version info ('Current baseline (checked 2026-06-07): RDKit 2026.03.3') would be more token-efficient in a dedicated versioning/compatibility note rather than the narrative overview.	4 / 5
Actionability	Concrete executable install commands and specific pitfall fixes (DetectChemistryProblems(), AddHs()) are present, but the body itself is mostly navigational and offloads worked code to references/ and scripts/ rather than including copy-paste-ready examples inline.	4 / 5
Workflow Clarity	Capabilities are presented as a table and pitfalls as a checklist rather than a sequenced workflow; validation is mentioned ('Always validate molecules after parsing') but not embedded as explicit validate-then-proceed checkpoints for batch/destructive operations like sanitization or filtering.	3 / 5
Progressive Disclosure	Clear overview with well-signaled one-level-deep references (core_capabilities.md, workflows_and_best_practices.md, api_reference.md, descriptors_reference.md, smarts_patterns.md) and a scripts/ directory, all verified to exist, with an explicit clarification of which names are packages versus local bundled files.	5 / 5
	Total	16 / 20 Passed

Description

100%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 exemplary: it names concrete capabilities, includes the natural domain vocabulary, explicitly answers both what and when, and disambiguates from the closely related datamol skill.

Dimension	Reasoning	Score
Specificity	Lists multiple concrete actions across the domain — 'SMILES/SDF parsing', 'descriptors (MW, LogP, TPSA)', 'fingerprints', 'substructure search', '2D/3D generation', 'similarity, reactions', 'custom sanitization' — giving comprehensive coverage rather than vague language.	5 / 5
Completeness	Explicitly states what it does (parsing, descriptors, fingerprints, search, generation, similarity, reactions) and when to use it ('Use rdkit for advanced control, custom sanitization, specialized algorithms'), plus an explicit contrast to datamol for the simpler case.	5 / 5
Trigger Term Quality	Comprehensive coverage of the natural terms a cheminformatics user would say, including package names (rdkit, datamol), file/line formats (SMILES, SDF), and operation keywords (fingerprints, substructure search, sanitization, similarity).	5 / 5
Distinctiveness Conflict Risk	Occupies a clear niche (RDKit cheminformatics) with distinct triggers and an explicit disambiguation against the datamol wrapper, minimizing overlap with sibling skills.	5 / 5
	Total	20 / 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/rdkit/SKILL.md
Commit: 9e8b0cb

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.