Content
42%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
A well-organized rule index with a few strong executable code examples, but its value depends on external rule files and a compiled document that are absent, so progressive disclosure fails and most rules are not actionable in situ.
Suggestions
Provide the referenced bundle files (rules/*.md and ../R3F_BEST_PRACTICES.md) or remove the broken references and inline the essential rule detail so the skill is self-contained.
De-duplicate content between the priority table, quick-reference lists, and 'Critical Patterns' to tighten the token budget.
Add at least a minimal apply/verify flow (e.g., 'check r3f-perf before and after') so the reference doubles as a guided workflow.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The rule catalog is mostly lean one-liners without explaining R3F basics, but the priority table, quick-reference lists, and 'Critical Patterns' section repeat the same perf rules, adding redundancy that could be tightened. | 2 / 3 |
Actionability | 'Critical Patterns' provides several complete BAD/GOOD executable code blocks, but the bulk of the 70+ rules are one-line reminders whose detailed guidance lives in referenced rule files that do not exist, so the guidance is only partially copy-paste ready. | 2 / 3 |
Workflow Clarity | The priority table and 'When to Apply' give some structure, but as a reference catalog there is no sequenced multi-step process or checkpoints; the 'How to Use' step ('Read individual rule files') points at missing files. | 2 / 3 |
Progressive Disclosure | The body signals one-level-deep references ('rules/perf-never-set-state-in-useframe.md', '../R3F_BEST_PRACTICES.md') and there are no references/, scripts/, or assets/ bundles, so the referenced files do not exist — the claimed progressive disclosure is not actually delivered, leaving a monolithic inline catalog. | 1 / 3 |
Total | 7 / 12 Passed |