Content
65%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
The body is highly actionable with executable code throughout, but it is over-long and duplicates content already in the bundled references, weakening conciseness and progressive disclosure. Workflow validation checkpoints are present only implicitly.
Suggestions
Trim the inline capability sections into a concise overview and defer exhaustive examples to references/workflows.md and references/api_quick_reference.md, linking each section explicitly.
Add explicit validate→fix→retry feedback loops for batch/risky operations (e.g. check solution.status after optimize, re-validate sampled fluxes before reporting).
Replace prose that restates API behavior with one-line pointers to the API quick reference to reduce token load.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The ~450-line body is mostly executable code with little concept padding, but it substantially overlaps the bundled references and could be tightened by pushing per-capability detail into those files. | 2 / 3 |
Actionability | Every section provides complete, copy-paste-ready Python (e.g. load_model, model.optimize(), single_gene_deletion) with real imports and arguments, matching the fully-executable anchor. | 3 / 3 |
Workflow Clarity | Workflows 1–5 are sequenced and best practices mention checking status/validating, but there are no explicit validate→fix→retry feedback loops for batch operations like double deletions or sampling, so it sits at 2. | 2 / 3 |
Progressive Disclosure | The two referenced files (references/workflows.md, references/api_quick_reference.md) are real and one level deep, but the main SKILL.md is a monolithic wall of inline capability detail that largely duplicates those references rather than being a lean overview pointing to them. | 2 / 3 |
Total | 9 / 12 Passed |