drug-discovery

Pharmaceutical research assistant for drug discovery workflows. Search bioactive compounds on ChEMBL, calculate drug-likeness (Lipinski Ro5, QED, TPSA, synthetic accessibility), look up drug-drug interactions via OpenFDA, interpret ADMET profiles, and assist with lead optimization. Use for medicinal chemistry questions, molecule property analysis, clinical pharmacology, and open-science drug research.

Quality

82%

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

65%

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

A highly actionable body built from executable, copy-paste-ready API commands, with a sensible ADMET reference link. It is weakened by a small amount of persona fluff, missing validation checkpoints in the analytical workflow, and orphaned script files that duplicate the inline code instead of being referenced.

Suggestions

Drop the 'You are an expert pharmaceutical scientist…' persona opener and the 'world's largest open bioactivity database' parenthetical so every token earns its place.

Add an explicit validation checkpoint to the Reasoning Guidelines (e.g. 'verify the property values against a second source before flagging liabilities') to turn the sequence into a validate→fix→retry loop.

Reference scripts/chembl_target.py and scripts/ro5_screen.py from the relevant workflows instead of inlining equivalent bash, so the skill follows one-level-deep progressive disclosure rather than duplicating content.

Dimension	Reasoning	Score
Conciseness	The bulk is lean, executable curl/python blocks, but the persona opener 'You are an expert pharmaceutical scientist and medicinal chemist…' and the parenthetical 'the world's largest open bioactivity database' explain things Claude already knows; not at 3 because not every token earns its place, not at 1 because the content is overwhelmingly efficient code.	2 / 3
Actionability	Provides fully executable, copy-paste-ready bash commands hitting real API endpoints (ChEMBL, PubChem, OpenFDA, OpenTargets) with inline parsing; matches the anchor for fully executable code and specific examples; not at 2 because the guidance is complete rather than pseudocode.	3 / 3
Workflow Clarity	Five numbered workflows plus a numbered reasoning checklist give a clear sequence, but only basic empty-result checks exist and there are no explicit validate→fix→retry checkpoints for the multi-step analysis; not at 3 because checkpoints are missing/implicit, not at 1 because steps are clearly listed and sequenced.	2 / 3
Progressive Disclosure	references/ADMET_REFERENCE.md is clearly signaled one level deep, but scripts/chembl_target.py and scripts/ro5_screen.py are never linked from the body and the inline bash duplicates them — content that should be separate is inline; not at 3 because navigation to the scripts is missing, not at 1 because some structure and a real signaled reference exist.	2 / 3
	Total	9 / 12 Passed

Description

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.

A strong, third-person description that states concrete capabilities and gives an explicit 'Use for…' trigger clause covering natural domain terms. It is distinctive and complete, answering both what the skill does and when to invoke it.

Dimension	Reasoning	Score
Specificity	Lists multiple concrete actions — 'Search bioactive compounds on ChEMBL', 'calculate drug-likeness (Lipinski Ro5, QED, TPSA…)', 'look up drug-drug interactions via OpenFDA', 'interpret ADMET profiles' — matching the anchor for several specific concrete actions; it is not at 2 because it goes well beyond naming a domain plus a few actions.	3 / 3
Completeness	Explicitly answers both what (the enumerated capabilities) and when via an explicit 'Use for medicinal chemistry questions, molecule property analysis, clinical pharmacology, and open-science drug research' trigger clause; the Use clause is present so it is not capped at 2.	3 / 3
Trigger Term Quality	Includes natural domain terms a user would actually say — 'drug discovery', 'drug-likeness', 'drug-drug interactions', 'ADMET', 'medicinal chemistry questions', 'molecule property analysis', 'clinical pharmacology'; not at 2 because coverage is broad rather than missing common variations.	3 / 3
Distinctiveness Conflict Risk	A clearly defined pharma/cheminformatics niche (ChEMBL, OpenFDA, ADMET, lead optimization) with distinct triggers unlikely to overlap with general skills; not at 2 because it is far more specific than 'Works with document files'.	3 / 3
	Total	12 / 12 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
metadata_version	'metadata.version' is missing	Warning
metadata_field	'metadata' should map string keys to string values	Warning
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	13 / 16 Passed

Repository: NousResearch/hermes-agent
Path: optional-skills/research/drug-discovery/SKILL.md
Commit: 4c9628e

Reviewed: about 15 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.