rowan
Rowan is a cloud-native molecular modeling and medicinal-chemistry workflow platform with a Python API. Use for pKa and macropKa prediction, conformer and tautomer ensembles, docking and analogue docking, protein-ligand cofolding, MSA generation, molecular dynamics, permeability, descriptor workflows, and related small-molecule or protein modeling tasks. Ideal for programmatic batch screening, multi-step chemistry pipelines, and workflows that would otherwise require maintaining local HPC/GPU infrastructure.
63
77%
Does it follow best practices?
Impact
—
No eval scenarios have been run
Risky
Do not use without reviewing
Optimize this skill with Tessl
npx tessl skill review --optimize ./scientific-skills/rowan/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 an excellent skill description that clearly identifies a specialized computational chemistry platform, lists numerous concrete capabilities, and provides explicit trigger guidance. The domain-specific terminology serves double duty as both capability documentation and natural trigger terms. The description is comprehensive yet focused, making it easy for Claude to distinguish this skill from others.
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | The description lists numerous specific concrete actions: pKa prediction, macropKa prediction, conformer ensembles, tautomer ensembles, docking, analogue docking, protein-ligand cofolding, MSA generation, molecular dynamics, permeability, descriptor workflows. This is highly specific and comprehensive. | 3 / 3 |
Completeness | Clearly answers 'what' (cloud-native molecular modeling platform with Python API, listing specific capabilities) and 'when' ('Use for pKa and macropKa prediction...', 'Ideal for programmatic batch screening, multi-step chemistry pipelines'). The 'Use for' clause serves as an explicit trigger guidance. | 3 / 3 |
Trigger Term Quality | Includes many natural domain-specific keywords users would actually say: pKa, docking, conformer, tautomer, molecular dynamics, permeability, cofolding, MSA, medicinal chemistry, small-molecule, protein modeling, batch screening, HPC/GPU. These are terms a computational chemist would naturally use. | 3 / 3 |
Distinctiveness Conflict Risk | Highly distinctive niche in computational chemistry and molecular modeling. The specific domain terms (pKa, docking, cofolding, MSA, molecular dynamics) and the platform name 'Rowan' make it very unlikely to conflict with other skills. | 3 / 3 |
Total | 12 / 12 Passed |
Implementation
55%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
The skill is highly actionable with excellent executable examples, clear workflow sequencing, and comprehensive error handling. However, it is severely bloated — a monolithic ~600+ line document that tries to be both a quick-start guide and a complete API reference simultaneously. The lack of any progressive disclosure (no bundle files, no external references) means the entire content competes for context window space on every invocation, violating token efficiency principles.
Suggestions
Split the content into a concise SKILL.md overview (~100-150 lines covering quick start, core pattern, and workflow decision trees) with references to separate files like WORKFLOWS.md (detailed workflow examples), WEBHOOKS.md (webhook setup and handler code), and REFERENCE.md (pricing, descriptor keys, all workflow function tables).
Remove the pricing/credit tables and cost estimates entirely or move them to a separate PRICING.md — this information is volatile and not needed for Claude to execute workflows correctly.
Eliminate redundant sections: the 'Quick start' and 'Core usage pattern' sections overlap significantly; the webhook secret management appears twice (in 'User and webhook management' and again in 'Webhooks and asynchronous workflows').
Trim explanatory text that Claude already knows (e.g., what SMILES/InChI are, what MSA is, general webhook best practices like 'respond quickly' and 'implement idempotency') to focus only on Rowan-specific API details.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | This is extremely verbose at ~600+ lines. It includes extensive pricing tables, credit cost estimates, webhook handler implementations, full FastAPI examples, decision trees, and exhaustive workflow catalogs that could easily be split into reference files. Much of this (e.g., explaining what MSA is, webhook best practices, protein preparation guidance) is knowledge Claude already has or could be dramatically condensed. | 1 / 3 |
Actionability | The skill provides fully executable, copy-paste-ready code examples for every workflow type, with correct parameter names, typed result access patterns, error handling, and a complete end-to-end lead optimization campaign. Code is concrete and specific with real SMILES strings and expected outputs. | 3 / 3 |
Workflow Clarity | Multi-step workflows are clearly sequenced (submit → wait → retrieve), decision trees help choose between workflow types, the end-to-end example demonstrates chaining (tautomer → pKa → descriptors → docking), and error handling/recovery patterns are explicit with try/except blocks and status checking. | 3 / 3 |
Progressive Disclosure | Everything is crammed into a single monolithic file with no bundle files or external references. The pricing tables, full webhook handler implementations, exhaustive workflow catalogs, and detailed API reference tables should be split into separate reference files. There is no progressive disclosure structure whatsoever. | 1 / 3 |
Total | 8 / 12 Passed |
Validation
72%Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.
Validation — 8 / 11 Passed
Validation for skill structure
| Criteria | Description | Result |
|---|---|---|
skill_md_line_count | SKILL.md is long (1088 lines); consider splitting into references/ and linking | Warning |
metadata_version | 'metadata.version' is missing | Warning |
metadata_field | 'metadata' should map string keys to string values | Warning |
Total | 8 / 11 Passed | |
- Repository
- K-Dense-AI/claude-scientific-skills
- Commit
cbcae7b
- 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.