Content
63%Weight 40%Scale 1-5Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
A highly actionable, well-structured body with executable code and embedded validation/troubleshooting guidance. Its two weaknesses are the ~470 lines of inline implementation code that could be trimmed, and the complete failure to reference the four existing scripts/ modules that duplicate that code — leaving the bundle orphaned and wasting the skill's progressive-disclosure architecture.
Suggestions
Replace the six inline capability implementations with brief usage notes plus pointers to the corresponding bundle files, e.g. "Full implementations: scripts/codon_optimize.py, scripts/gene_circuit.py, scripts/sbml_model.py, scripts/bifurcation.py" — this both fixes the orphaned bundle and cuts ~470 lines of body code.
Move the 23-line inline E. coli codon usage table into a script or reference file, keeping only a one-line usage example in SKILL.md.
Make each snippet self-contained: import pandas inside sensitivity_analysis and define or reference toggle_switch explicitly in Workflow 2.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The prose is lean and assumes Claude's domain competence, but ~470 lines of implementation code are inlined — including a 23-line E. coli codon table and full ODE/SBML/fitness implementations — that duplicate the 2,221-line scripts/ bundle, so the body could be substantially tightened. It is not 2 because there is no padding with concepts Claude already knows; the excess is structural (penalized under progressive_disclosure) rather than explanatory. | 3 / 5 |
Actionability | A runnable Quick Start plus six largely copy-paste-ready capability blocks, three worked workflows, an install command, and a troubleshooting section give concrete executable guidance. It is not 5 because of minor gaps: sensitivity_analysis uses pd.DataFrame without importing pandas in that snippet, and Workflow 2 depends on toggle_switch being defined in Quick Start. | 4 / 5 |
Workflow Clarity | Typical Workflows give concrete end-to-end sequences, Best Practices embed checkpoints ("always call doc.getNumErrors() after model creation", "always check GC content after optimization"), and Troubleshooting provides fix-retry feedback (e.g. "excess work" → switch to BDF/Radau). It is not 5 because the workflows read as worked code examples rather than step-numbered procedures with explicit validate-before-proceed gates. | 4 / 5 |
Progressive Disclosure | Section structure is well-organized (Overview / When to Use / Capabilities / Workflows / Best Practices / Troubleshooting / Resources), but the four scripts/*.py bundle files (2,221 lines) are never referenced anywhere in the body — all of their functionality is re-inlined instead — matching the anchor "content that should be separate is inline" with references absent. It is not 2 because the structure itself is solid rather than minimal; it is not 4 because zero signaling of an existing bundle is more than a minor organization gap. | 3 / 5 |
Total | 14 / 20 Passed |