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.
The body is a well-organized, expert-level compliance knowledge base with genuinely actionable decision frameworks, escalation triggers, and metrics. Its weaknesses are token efficiency — substantial restatement of domain knowledge Claude already has — and the absence of any progressive disclosure: everything is inlined in a single long file rather than split into reference bundles.
Suggestions
Split reference-grade material (GRI rule walkthrough, Incoterms definitions, regional specialties, edge cases) into one-level-deep files under references/ and keep SKILL.md as an overview with clearly signaled links, e.g. 'Incoterms compliance details: See [INCOTERMS.md](references/INCOTERMS.md)'.
Trim the textbook explanations of concepts Claude already knows (Incoterms definitions, GRI basics, what the HS is) down to only the non-obvious compliance implications — the pitfalls, valuation circularity, and misunderstanding callouts.
Expand the 'Key Templates' section into fully copy-paste-ready skeletons (complete subject lines, body structure, required fields) instead of brief sketches, or move them to a references file where full versions can live without bloating the main file.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is dense rather than padded, but several sections restate knowledge Claude already has — textbook Incoterms definitions ('EXW: Seller's minimum obligation. Buyer arranges everything'), the GRI walkthrough with standard examples ('A car body without the engine is still classified as a motor vehicle'), and 'The Harmonized System is a 6-digit international nomenclature maintained by the WCO.' Not 2: nearly every restatement is compressed and carries a compliance-specific caveat; not 4: the Incoterms and GRI sections could be trimmed to only the non-obvious compliance implications. | 3 / 5 |
Actionability | Concrete, executable guidance throughout: numbered decision frameworks ('Matches below 85% name similarity with no address or country correlation are likely false positives — document and clear'), penalty formulas, an escalation table with timelines, and message templates. Not 5: the templates are explicitly brief sketches to adapt rather than copy-paste ready, and there is no executable code or commands. | 4 / 5 |
Workflow Clarity | The multi-step processes ('Classification Decision Logic', 'FTA Qualification Analysis', 'Valuation Method Selection', 'Screening Hit Assessment') are clearly sequenced with checkpoints such as 'Validate at the subheading level' and 'Check for binding rulings.' Not 5: there are no explicit validate-then-retry feedback loops for error recovery, though this is an advisory skill rather than a destructive/batch one. | 4 / 5 |
Progressive Disclosure | No bundle files exist and the skill is a ~240-line monolith that inlines reference-grade material — regional specialties, eight edge cases, penalty frameworks, templates — that belongs in separate one-level-deep reference files. Not 2: sections are well-organized with clear headers and the body explicitly frames itself as summaries to expand; not 4: nothing is split out of SKILL.md, and 'Additional Resources' points to internal artifacts rather than actual bundle files. | 3 / 5 |
Total | 14 / 20 Passed |