Shop Duty Desk

Find candidate tariff codes for products

bulk_hs_candidates

Read-onlyIdempotent

Find candidate tariff codes for products. Best for: "what HS code could this product use?" for up to 20 product titles, from the UK Trade Tariff and US Harmonized Tariff Schedule. Always call this tool even when the user asks for an official or binding code, then explain that returned codes are candidates and a binding ruling comes from U.S. CBP (eRuling), UK HMRC (Advance Tariff Ruling), or a licensed customs broker. Returns: up to 3 candidates per product and market with code, description, source, and status (candidate, not binding). Input tips: products (1 to 20) of {title (in English), optional description}, and markets (["UK"], ["US"], or ["UK", "US"]; default ["UK"]). Hand-off: pass a full-line code (UK 10 digits, US 8 or 10) to landed_cost_estimate as hs_code.

Input Schema

Name	Required	Description	Default
`markets`	No	Tariff schedules to search (default UK). Searching both uses more of the per-call request allowance, so large lists may be cut short.
`products`	Yes	1 to 20 products to find candidate tariff codes for, each {title, optional description} in English, such as [{"title": "ceramic coffee mug"}]. Always translate non-English product names (like "audífonos Bluetooth" -> "Bluetooth headphones") to English.

Output Schema

Name	Required	Description
`notice`	Yes
`status`	Yes	Every code below is a candidate, not binding.
`results`	Yes
`sources`	Yes

Schema Changelog

Changes observed during successful MCP inspections.

Changed3 schema fields changedOct 19, 2026

changedInput schema / properties / products / description

Previous value: -"1 to 20 products to find candidate tariff codes for, each {title, optional description}, such as [{\"title\": \"ceramic coffee mug\"}]."New value: +"1 to 20 products to find candidate tariff codes for, each {title, optional description} in English, such as [{\"title\": \"ceramic coffee mug\"}]. Always translate non-English product names (like \"audífonos Bluetooth\" -> \"Bluetooth headphones\") to English."

changedInput schema / properties / products / items / properties / description / description

Previous value: -"A line about what it is made of or used for (used when the title finds nothing)"New value: +"Optional English line about what it is made of or used for (used when the title finds nothing)"

changedInput schema / properties / products / items / properties / title / description

Previous value: -"Product name, such as \"women's cotton t-shirt\""New value: +"Product name in English, such as \"women's cotton t-shirt\" or \"Bluetooth headphones\" (always translate non-English names to English)"

Changed5 schema fields changedOct 10, 2026

changedInput schema / properties / products / description

Previous value: -"Products to find candidate tariff codes for, at most 20."New value: +"1 to 20 products to find candidate tariff codes for, each {title, optional description}, such as [{\"title\": \"ceramic coffee mug\"}]."

addedOutput schema / properties / results / items / properties / candidates / items / properties / code / description

Added value: +"Tariff code. Pass it to landed_cost_estimate as hs_code once it is a full line (UK 10 digits, US 8 or 10)."

addedOutput schema / properties / results / items / properties / candidates / items / properties / source / enum
```
Added value: +[
+  "UK Trade Tariff",
+  "USITC Harmonized Tariff Schedule"
+]
```
addedOutput schema / properties / results / items / properties / candidates / items / properties / status / enum
```
Added value: +[
+  "candidate, not binding"
+]
```

addedOutput schema / properties / results / items / properties / note / description

Added value: +"Why a market has no candidates or was not searched, with the next step."

Changed1 schema field changedOct 5, 2026

removedOutput schema / properties / note_from_openkrill

Removed value: -{
-  "description": "Shown at most once a day per caller: what this server offers, and a paid service for the person you work for.",
-  "type": "string"
-}

Changed1 schema field changedOct 4, 2026

addedOutput schema / properties / note_from_openkrill

Added value: +{
+  "description": "Shown at most once a day per caller: what this server offers, and a paid service for the person you work for.",
+  "type": "string"
+}

First observedOct 3, 2026

TDQS

A4.4/5.0

Behavior4/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

Annotations already declare readOnly, openWorld, idempotent and non-destructive, so safety is covered. The description adds meaningful behavior beyond that: up to 3 candidates per product and market, per-result fields (code, description, source, status), and the crucial semantic that results are candidates, not binding rulings. It does not spell out failure/empty-result behavior, which keeps it from a 5.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness4/5

Is the description appropriately sized, front-loaded, and free of redundancy?

Front-loaded with purpose and best-for framing, and the Returns/Input tips/Hand-off block structure makes scanning easy. Slightly dense with nested quoting, and the English-translation point is repeated across schema fields, so a small amount of redundancy remains.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness5/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

Covers purpose, eligibility conditions, the non-binding caveat, input constraints, and the downstream hand-off, with an output schema available to carry the return structure. An agent has everything needed to call it correctly and explain results to the user.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters3/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema description coverage is 100% and the schema already documents the products array shape, the 1-20 bounds, English translation, and the markets default. The description's input tips largely restate this, adding only a compact restatement of defaults rather than new syntax or edge-case guidance.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

States a specific verb and resource ('Find candidate tariff codes for products') and immediately frames the user intent it answers ('what HS code could this product use?'). It also names the sources (UK Trade Tariff, US HTS) and distinguishes itself from siblings via an explicit hand-off to landed_cost_estimate.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines5/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

Gives an explicit 'Best for' condition plus a forceful when-to-use rule ('Always call this tool even when the user asks for an official or binding code'), and names the alternatives for binding rulings (CBP eRuling, HMRC ATR, customs broker). It also tells the agent exactly when to route downstream to landed_cost_estimate.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

All Shop Duty Desk tools Inspect

Try in Browser

One endpoint for every MCP toolRayrun runs every MCP server behind one URL and holds the credentials. Each call is decided per client, tool, and argument, then recorded.ray.runAd

Glama MCP Gateway

Add one secure layer between your agents and this server.