Skills
Customs Trade Compliance Skill Review: Installation
The customs-trade-compliance skill packages HS classification rules, FTA evaluation, and denied-party screening logic into Claude Code. Zero installs and no ratings signal it is untested at production scale; the SKILL.md is thorough but requires careful trigger wording to avoid misrouting.
An agentskill is a pre-written package of role context, domain knowledge, and procedures that a developer or operator can load into Claude to handle a specific class of task. The customs-trade-compliance skill is one such package, authored by affaan-m, that aims to teach Claude the language, rules, and workflows of international trade compliance.
Marketplace Snapshot
| Metric | Value |
|---|---|
| agentskill rating | 0 (no ratings) |
| Total installs | 0 |
| Security score | 100/100 |
| Content quality | 100/100 |
| Last updated | 2026-05-19 |
| GitHub stars (upstream repo) | 185,629 |
The numbers here are clear: zero installs, zero user ratings, full security and content marks from automated checks. The skill is new. It has not been battle-tested in production. This is the critical fact to hold: everything else in the review assumes you understand that no real operators have used it yet, so you are not inheriting anyone’s hard-learned edge cases.
What the Skill Declares
The SKILL.md opens with a role statement: “You are a senior trade compliance specialist with 15+ years managing customs operations across US, EU, UK, and Asia-Pacific jurisdictions.” It then specifies when to apply the skill: classifying goods under HS/HTS codes, preparing customs documentation, screening parties against denied-party lists, evaluating FTA qualification, and responding to audits.
The core content is organized into six sections: role and context, trigger conditions, how-it-works steps, worked examples, core knowledge (tariff classification and documentation), and a reference for common mistakes.
Installation and Access Path
To use this skill in Claude Code, you would download it from the agentskill.sh marketplace and place it at .claude/skills/affaan-m/customs-trade-compliance/SKILL.md in your Claude workspace. Once installed, you invoke it by naming it in a prompt fragment or by configuring it as an auto-load skill that Claude checks before every run.
The marketplace listing is at https://agentskill.sh/affaan-m/customs-trade-compliance. The upstream source repository is https://github.com/affaan-m/everything-claude-code, which has 185,629 stars on GitHub, suggesting a large and active collection of skills from the same author.
What It Actually Covers
The SKILL.md devotes the most space to HS tariff classification, walking through the six General Rules of Interpretation (GRI) in order and explaining when each applies. GRI 1, for example, resolves about 90% of classifications by reading heading text and checking Section and Chapter notes. GRI 3(a) applies when goods are prima facie classifiable under two or more headings, preferring the most specific. The skill includes examples of common misclassifications: multi-function devices, food preparations versus ingredients, textile composites, and parts versus accessories.
Documentation is covered with reference to US CBP rules (19 CFR, 19 USC), UK customs (CHIEF/CDS), and EU systems (TARIC). The skill describes commercial invoices, packing lists, certificates of origin, bills of lading, ISF 10+2 (the US Importer Security Filing), and entry summary forms. Each section specifies what must appear on the document and which CBP or regulatory citation applies.
The skill also outlines FTA evaluation (USMCA regional value content, tariff shift rules) and denied-party screening against the SDN list, Entity List, and EU sanctions.
What It Does Not Cover
The SKILL.md does not provide live tariff rate databases. You cannot ask Claude, using this skill, “What is the current HTS code 8542.31.00 duty rate?” and expect a real-time answer. The skill teaches you how to classify, not what the rates are. This is a significant limitation for operators who need to quote landed costs or assess duty exposure quickly.
The skill also does not integrate with ACE, CHIEF/CDS, ATLAS, or any real customs system. It provides examples of how those systems work but does not call them. If your workflow requires pulling current ISF data from ACE or filing entry summaries electronically, you would need a separate integration layer.
Denied-party screening is described in procedure but not automated. The skill walks you through how to respond to a screening hit, but it does not fetch the latest SDN list from OFAC or cross-reference your customer master data. You would use this skill to coach Claude on what to do after a hit, not to perform the initial screening.
Finally, the skill does not address post-entry compliance, supply-chain visibility systems, or landed-cost accounting. It focuses on the customs-clearance phase, not the broader trade management lifecycle.
Comparison with Baseline Claude and Alternatives
Claude 3.5 Sonnet, out of the box, has broad knowledge of HS codes, tariff structure, and trade regulations. It can classify many simple products correctly and explain GRI rules. However, it is slow to apply GRI rules in strict order, sometimes conflates tariff terminology across jurisdictions, and may miss nuances in FTA qualification (e.g., regional value content calculations). The customs-trade-compliance skill offloads this burden by providing a condensed reference that Claude can cite and follow without having to regenerate the logic each time.
| Capability | Baseline Claude 3.5 Sonnet | customs-trade-compliance Skill | Custom In-House Skill |
|---|---|---|---|
| GRI rule application | General knowledge, no strict order | Explicit step-by-step GRI 1-6 | Org-specific precedents, case law |
| Documentation templates | Partial, generic | HS, FTA, ISF, entry forms | Org legal language, EDI schema |
| FTA qualification logic | Basic USMCA awareness | USMCA RVC, tariff shift, Chapter Rules | Org product families, suppliers |
| Denied-party screening | Conceptual only | Procedure for hits and escalation | Live SDN API, compliance dashboard |
| Tariff rate lookup | Outdated or encyclopedic | None (reference only) | Real-time HTS.gov or TARIC API |
| Audit response strategy | Generic | Penalty mitigation framework | Org enforcement history, counsel briefs |
A specialized in-house skill would encode your organization’s product mix, preferred FTA programs, and legal counsel’s red lines. This general-purpose skill is most useful if you lack internal expertise or need a starting checklist.
Strengths and Failure Modes
Strengths:
The SKILL.md is well-structured and comprehensive for its scope. The GRI rules are explained clearly with examples, reducing the chance that Claude will skip a required step. The documentation section cites actual regulations (19 CFR, 19 USC) and actual forms (ISF, entry summary), which is rare in public AI resources. The worked examples (HS classification dispute, FTA qualification, denied-party hit) are realistic and walk through reasoning, not just conclusions.
Failure modes:
First, trigger wording matters. If you ask “Is this product dutiable?” without naming the skill explicitly or configuring auto-trigger, Claude may not invoke the skill’s role context and will revert to baseline knowledge. The marketplace does not show whether auto-trigger is configured for this skill or what keywords activate it.
Second, zero installs means zero feedback loops. You do not know whether the skill’s GRI ordering holds up under pressure, whether its documentation templates pass CBP scrutiny, or whether its FTA logic is sound for edge cases (e.g., short-supply exemptions, trade remedies). The first user will be debugging in production.
Third, the skill provides no way to flag when it is out of date. Tariff schedules change, sanctions lists grow, and trade agreements are renegotiated. A skill installed on 2026-05-19 may carry advice that was correct on that date but wrong six months later. The SKILL.md does not include a “last validated” date for each fact or a process for updating the skill when regulations shift.
Fourth, the skill’s denial-party screening section is procedural but not actionable without a live API. It tells you to “screen all transaction parties,” but it does not tell you how to do it at scale or how to track false positives. For a small importer with a handful of shipments per month, this is fine; for a high-volume distributor, it leaves the hard part unsolved.
When to Use This Skill
Use this skill if:
You are a small to mid-size importer or exporter without a dedicated trade compliance officer and want Claude to help you understand HS classification rules before you hire a customs broker or file an entry.
You are a logistics provider or freight forwarder and want to coach Claude on documentation requirements so it can validate invoices and BOLs before they reach customs.
You are a compliance officer building an internal checklist and want to start with a public reference frame rather than blank paper.
You are experimenting with agentic workflows and want to see how domain-specific knowledge can be packaged as a reusable skill.
Skip this skill if:
You are a high-volume importer with regulatory obligations and penalties in the millions of dollars. Use a licensed customs broker or a commercial trade management system with live APIs and audit trails.
You need real-time tariff rate lookup, landed-cost calculation, or integration with your ERP. This skill does not provide those functions.
You operate in a jurisdiction not covered by the SKILL.md (China, Southeast Asia, etc.). The skill focuses on US, EU, UK, and Asia-Pacific generically; it may not reflect local practices.
You have active customs disputes or audit defense work. The skill’s penalty mitigation section is a framework, not a defense strategy. Work with counsel.
Takeaways
The customs-trade-compliance skill is thorough within its scope and provides value as a training reference or starting point. The GRI rules, documentation templates, and worked examples are correct and well-explained. However, zero installs and zero ratings mean it has not been tested in production; you would be the first adopter.
The skill excels at teaching the rules but does not automate the tasks. You cannot ask it to file an ISF, fetch a live tariff rate, or screen a customer against OFAC. It teaches Claude to coach you through those steps.
Tariff and sanctions regulations change frequently. This skill carries a creation date but no maintenance schedule. Before using it in production, verify that its references to current HTS codes, FTA rates, and sanctions lists are still accurate.
Use this skill as a reference layer for baseline knowledge or as a starting template for a custom in-house skill. Do not rely on it alone for regulated exports, high-value shipments, or jurisdictions where penalties are steep.
Further reading
- agentskill.sh listing for customs-trade-compliance: The official marketplace page with version history and user feedback submission interface.
- GitHub repo: everything-claude-code: The upstream source repository containing this and related Claude Code skills.
- WCO Harmonized System nomenclature: The international tariff classification standard that the skill’s GRI rules implement.
- US CBP Tariff Resources (HTS): The authoritative source for current US HTS codes and duty rates.
- USMCA Rules of Origin: Detailed guidance on regional value content and tariff shift qualifying for the USMCA agreement mentioned in the skill.
Frequently asked
What does the customs-trade-compliance skill actually do?
It provides Claude with role context for a trade compliance specialist, including HS/HTS tariff classification logic (the GRI rules in strict order), documentation templates for commercial invoices and certificates of origin, FTA qualification frameworks (e.g. USMCA regional value content), and denied-party screening procedures. It is not a live tariff database or a real-time API integration; it is reference knowledge bundled as context.
Can I use this skill without custom code?
Yes. Once installed, the skill provides structured prompt context that Claude can reference. You invoke it by naming it in the prompt (e.g., 'Using the customs-trade-compliance skill, classify this product') or by configuring it as an auto-trigger. You do not need to write parsing or integration code unless you want to pipe results into an ERP system.
How current are the tariff rates and regulations in this skill?
The SKILL.md does not list a tariff rate database or a last-updated date for duty schedules. The skill provides classification methodology (GRI rules, chapter structures) but relies on the operator to verify current rates against official sources like HTS.gov or TARIC. For live tariff queries, you would need a separate API integration or lookup tool.
Why has this skill had zero installs?
The skill was listed on agentskill.sh on 2026-05-19 and has not yet been adopted. Trade compliance is a specialized domain with a small practitioner base, and practitioners may not yet be aware of the skill or may prefer proprietary platforms. No installs does not indicate a quality defect; it indicates the feature is new and unproven in production.
What platforms does this skill support?
The marketplace listing shows platforms as 'unknown,' but the skill is written as Claude Code, which means it integrates with Claude 3.5 Sonnet and later via the Claude Code execution environment. It does not natively integrate with Zapier, Make, or other low-code platforms unless you add a separate bridge.