datamol

A Pythonic wrapper around RDKit with simplified interfaces and sensible defaults. Preferred for standard drug discovery workflows including SMILES parsing, standardization, descriptors, fingerprints, clustering, 3D conformer generation, and parallel processing. Returns native rdkit.Chem.Mol objects. For advanced control or custom parameters, use rdkit directly.

Quality

60%

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

Fix and improve this skill with Tessl

tessl review fix ./scientific-skills/Other/datamol/SKILL.md

Quality

Content

50%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 rich with concrete, executable datamol examples and has clearly signaled one-level-deep reference files, but it is severely padded with generic boilerplate and a promotional block, inlines ~600 lines of API reference that belongs in the reference files, and lacks explicit validation feedback loops for its batch workflows. It is most constrained by conciseness and by the inlining that weakens progressive disclosure.

Suggestions

Move the bulk of the inline API reference (Core Workflows sections 1-10, Best Practices, Error Handling, ML Integration) into the existing references/ files, keeping only a concise Quick Start and pointers in the body to reduce ~780 lines toward a lean overview.

Remove the generic boilerplate (restated description in Key Features, "Implementation Details" that points back to Overview) and the promotional "Use K-Dense Web" block that add no execution value.

Add explicit validate→fix→retry checkpoints to the batch pipelines (e.g., check parsed mols are not None / DataFrame non-empty before descriptor computation, with a retry-on-failure step) to raise workflow clarity above the batch cap of 3.

Dimension	Reasoning	Score
Conciseness	The ~780-line body has several padded/generic boilerplate sections ("When to Use", "Key Features" that restates the description verbatim, "Implementation Details" pointing back to Overview, and a promotional "Use K-Dense Web" block) plus inline API reference that duplicates the reference files, matching the 2 anchor's "several unnecessary explanations or padded sections" rather than the mostly-efficient 3.	2 / 5
Actionability	Dense, mostly executable examples (dm.to_mol, dm.read_sdf, dm.descriptors.batch_compute_many_descriptors, full Load→Filter→Analyze pipelines) cover common cases, with only minor gaps such as the placeholder `dm.operation(..., n_jobs=-1)` and code blocks that often lack ```python fences.	4 / 5
Workflow Clarity	Sequenced multi-step pipelines exist (numbered Load→Filter→Analyze steps, Recommended Workflow), but this batch-oriented skill lacks explicit validate→fix→retry feedback checkpoints in its pipelines, which the guidelines cap at 3 for batch operations.	3 / 5
Progressive Disclosure	References are clearly signaled (a Reference Documentation section listing all six real references/ files plus inline "Refer to references/X.md" pointers), but the bulk of the API reference is inlined in the body rather than placed in those separate files, matching the 3 anchor ("content that should be separate is inline") rather than the 4's "bulk in separate file".	3 / 5
	Total	12 / 20 Passed

Description

71%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 gives a strong, specific account of what the skill does and a clear niche boundary, but it lacks an explicit "Use when..." trigger clause, capping completeness at 3 and leaving trigger-term coverage just short of comprehensive. It is well-targeted to cheminformatics users but would benefit from explicit user-facing trigger phrasing.

Suggestions

Add an explicit "Use when..." trigger clause naming natural user phrases (e.g., "Use when parsing or standardizing SMILES, computing molecular descriptors/fingerprints, clustering molecules, or generating 3D conformers").

Include common synonyms and file extensions (molecules, .sdf, .smi, .xlsx) in the description to raise trigger-term coverage toward comprehensive.

Tighten the "Preferred for standard drug discovery workflows" phrasing into a direct trigger so the "when" is explicit rather than implied.

Dimension	Reasoning	Score
Specificity	Lists multiple concrete capabilities ("SMILES parsing, standardization, descriptors, fingerprints, clustering, 3D conformer generation, and parallel processing") plus a concrete return type ("Returns native rdkit.Chem.Mol objects") and a usage boundary, giving comprehensive coverage rather than the 1-2 actions of a 3.	5 / 5
Completeness	The "what" is clear and concrete, but the "when" is only weakly implied via "Preferred for standard drug discovery workflows including..." with no explicit "Use when..." trigger clause, which the guidelines cap at 3.	3 / 5
Trigger Term Quality	Good domain keyword coverage with natural cheminformatics terms (SMILES, descriptors, fingerprints, clustering, conformers, drug discovery), but lacks synonyms and file extensions (no .sdf/.smi, no generic "molecules") that the 5 anchor requires.	4 / 5
Distinctiveness Conflict Risk	A clear RDKit/cheminformatics niche with an explicit boundary ("For advanced control or custom parameters, use rdkit directly") minimizes conflict, but the absence of crisp explicit trigger phrases leaves minor overlap risk versus a raw RDKit skill rather than the minimal-conflict 5.	4 / 5
	Total	16 / 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 (791 lines); consider splitting into references/ and linking	Warning
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning
relative_links	Relative link issues: 2 missing	Warning

	Total	13 / 16 Passed

Repository: aipoch/medical-research-skills
Path: scientific-skills/Other/datamol/SKILL.md
Commit: f5ef65b

Reviewed: ~14 hours 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.