Content
47%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
The body delivers useful tool-selection and verification guidance with clear reasoning workflows, but spends most of its tokens restating neuroscience concepts Claude already knows and offers no executable code despite a Python mandate.
Suggestions
Trim the textbook reference sections (Nernst/Goldman equations, action-potential phases, STDP/mean-field theory, cranial-nerve and UMN/LMN tables) that Claude already knows, keeping only the look-up-first directive, tool table, and pitfalls.
Add at least one executable Python example for the 'ALWAYS use Python' quantitative mandate, and show a sample tool invocation (e.g. a PubMed_search_articles call with arguments) so guidance is copy-paste ready.
Move the computational-neuroscience equations and clinical-neurology reference into separate reference files linked from a concise overview, so the main SKILL.md stays a lean entry point.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The bulk is textbook neuroscience Claude already knows (rate-based models, Nernst/Goldman equations, action-potential phases, STDP, mean-field theory, cranial-nerve list, UMN/LMN table), padded as inline reference material rather than adding only what Claude doesn't know. | 1 / 3 |
Actionability | Provides a concrete tool table with parameters and copy-paste-ready search-query templates ('[region A] projection [region B]'), but mandates 'ALWAYS use Python' with no executable code examples and shows no example tool invocations, leaving key details missing. | 2 / 3 |
Workflow Clarity | Reasoning patterns (quantitative strategy, 'where in the brain', clinical neuro, tool selection) are clearly numbered with explicit VERIFY checkpoints, and the 'Common Pitfalls' section serves as a checklist for error-prone cases. | 3 / 3 |
Progressive Disclosure | Well-organized into seven numbered sections, but with no bundle files the single ~185-line SKILL.md is monolithic; detailed reference blocks (computational equations, clinical neurology, tool catalog) are inlined where separate files would aid navigation. | 2 / 3 |
Total | 8 / 12 Passed |