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 knowledge-rich, well-structured advisory skill with genuinely specific regulatory guidance and clear decision frameworks. Its weaknesses are verbosity (role padding and background explanations Claude already knows) and the absence of any progressive disclosure — all reference-grade material lives inline in a single large file.
Suggestions
Split reference-grade material (Regional Specialties, Key Edge Cases, Communication Patterns, Performance Indicators) into files under references/ and keep SKILL.md as a lean overview with well-signaled one-level-deep pointers.
Delete the 'Role and Context' persona paragraph and the definitional asides (what the HS is, what a BOL is, that Incoterms are not law), keeping only the compliance-relevant implications.
Promote the 'Key Templates' element lists into one fully worked example each (e.g., a complete prior-disclosure letter skeleton) so the guidance is copy-paste ready.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is dense with genuinely specialized data (penalty multipliers under 19 USC § 1592, RVC formulas, ISF data elements, drawback rules), but it also pads with a role-playing paragraph ('a senior trade compliance specialist with 15+ years') and explains concepts Claude already knows ('The Harmonized System is a 6-digit international nomenclature maintained by the WCO', 'Incoterms... are not law'). This matches the 3 anchor ('mostly efficient but includes some unnecessary explanation'); it is not a 2 because the majority of the content is non-generic regulatory detail that Claude cannot reliably recall. | 3 / 5 |
Actionability | Concrete guidance dominates: ordered decision frameworks, specific statutes and thresholds ('Matches below 85% name similarity... are likely false positives'), executable formulas (RVC = ((TV - VNM) / TV) × 100), escalation timelines, and template structures. It falls short of 5 because the communication templates are element lists ('Include: entry numbers, dates, specific violations...') rather than copy-paste-ready text, and no complete worked example (e.g., a filled prior disclosure) is provided. | 4 / 5 |
Workflow Clarity | Multi-step processes are clearly sequenced (7-step classification logic, 6-step FTA qualification, 5-step screening protocol) with checkpoints ('Validate at the subheading level', 'Never proceed with a transaction while a screening hit is unresolved', 'Document the rationale'). It is not a 5 because there are no explicit validate-fix-retry feedback loops, and the sequence in 'How It Works' duplicates rather than references the detailed frameworks. | 4 / 5 |
Progressive Disclosure | No bundle files exist (no references/, scripts/, or assets/), and the ~240-line body inlines core knowledge, regional specialties, edge cases, communication templates, and metrics tables that would serve better as separate reference files. Section headers are well-organized, which keeps it at 3 ('some structure but... content that should be separate is inline') rather than 2, but a skill this size offers no progressive disclosure at all — everything loads into context on every invocation. | 3 / 5 |
Total | 14 / 20 Passed |