Content
43%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.
The body is well-organized and rich in concrete rules, but it reads as a template plus rule catalog rather than an operational skill: no sequenced workflow connects repo source to finished README, key tooling references (nocodaigo, godocai scan) are unexplained, and long rule catalogs that belong in reference files are inlined. It scores slightly below the rubric midpoint overall.
Suggestions
Add an explicit numbered workflow (inspect repo source → collect the input table fields → draft sections per template → apply anti-slop rules → verify facts/links) with a validation checkpoint before finalizing.
Move the 12 anti-pattern catalog, banned-words list, and formatting rules into a references/ file (e.g. references/anti-slop.md), keeping only a summary in SKILL.md.
Explain or operationalize the repo-analysis step: how placeholders like <descrizione architettura> get filled and what "nocodaigo + godocai scan" invocations look like, and strip project-specific noise (hardcoded dates, personal links) from the template.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The rule catalogs are dense and skill-specific (not concepts Claude already knows), but the ~60-line README template embeds unnecessary project-specific noise ("https://github.com/bidoia", "aigoproxy · goleciave · PebbleDB", "Go 1.27.1", repeated "2026-11-20" dates) that could be trimmed, and the time-sensitive dates penalize conciseness per the rubric guidelines. This fits 'Mostly efficient but includes some unnecessary explanation or could be tightened' rather than 4's minor trims. | 3 / 5 |
Actionability | The input table, full output template, and rules with concrete values (#09090b, 127.0.0.1, specific banned words) provide real guidance, but the core transformation — how to derive the README content from the repo source — is never operationalized: placeholders like "<descrizione architettura>" and tools like "nocodaigo + godocai scan" are mentioned with no instruction on filling or running them. This matches 'Some concrete guidance but incomplete' rather than 4's 'minor gaps'. | 3 / 5 |
Workflow Clarity | There is no sequenced process at all: the implied flow (analyze repo → gather inputs → apply template → apply rules → verify) must be inferred from section order, steps are poorly defined, and validation is absent as a process step ("Verifica" appears only as text inside the output template). This fits 'Rough sequence present but many gaps; validation absent' rather than 3, which requires steps actually listed. | 2 / 5 |
Progressive Disclosure | Sections are clearly headed and the skill is self-contained with no bundle files, but at ~147 lines the 12-item anti-pattern catalog, banned-words list, formatting rules, and full README template are exactly the content that belongs in one-level-deep reference files. This matches 'Some structure but could be better organized; content that should be separate is inline' rather than 4, since the split would be substantial, not a minor gap. | 3 / 5 |
Total | 11 / 20 Passed |