HTS Classification Agent.
Find any US tariff code in seconds.

Describe a product. The agent searches CBP rulings, applies GRI rules, verifies against the live HTS schedule, and returns the 10-digit code with the actual duty rate importers paid (US Census Bureau). Web UI below — API access included.

134,050CBP Rulings
19,738HTS Codes
$4.8TTrade Data
CensusDuty Source
Complete the captcha first.
View raw JSON 

    

  







  

    
What the API returns

    

      

        
10-digit HTS codes

        
Up to 3 candidates with confidence levels and GRI-based rationale citing tariff schedule notes.

      

      

        
Effective duty rates

        
What CBP actually collected — from the US Census Bureau. Breakdown by MFN, Section 301/232, FTA, and exclusions.

      

      

        
CBP ruling evidence

        
Semantically matched from 134K CROSS rulings. Shown only when relevant to your product.

      

      

        
Legal notes

        
Official Section, Chapter, and Heading notes. The GRI rules customs brokers use to classify.

      

      

        
Progressive clarification

        
When uncertain, asks the specific question that resolves the next level in the tariff hierarchy.

      

      

        
Import volume

        
How much is imported, from where, at what rate. Real Census trade data.

      

    

  







  

    
One POST request

    
JSON in, JSON out. No SDK required.

    
# Classify a product
curl -X POST https://htsapi.dev/v1/classify \
  -H "Content-Type: application/json" \
  -H "X-API-Key: YOUR_API_KEY" \
  -d '{
    "description": "yoga mat",
    "country_of_origin": "China"
  }'

# Response includes:
# - hts_code: "9506.91.00.30" (high confidence)
# - census_duties: 38.3% effective from China
# - rulings: 3 CBP rulings confirming classification
# - legal_notes: Chapter 95 GRI notes
# - clarification: null (confident)

  







  

    
Guides

    

      
        
Developer Guide

        
Python, Node.js, curl — integrate in 5 minutes

      
      
        
Duty Calculator

        
Real effective rates from Census Bureau

      
      
        
Section 301 Tariffs

        
Actual duty collected on China imports

      
      
        
Importing from China

        
Duty rates, de minimis, cost reduction

      
      
        
FTA Savings

        
Preferential rates across 19 countries

      
      
        
Accuracy

        
70% at 10-digit on novel CBP rulings

      
    

    

      Also: Duty stacking · De minimis · CBP rulings · Audit response
    

  







  

    
How the HTS classification API works

    
An AI agent does the investigation. One API call returns the code with evidence.

    

      

        
1. CBP rulings, first

        
The agent searches 134,050 CBP precedents for similar products. When a ruling clearly applies, it follows authoritative precedent.

      

      

        
2. GRI reasoning

        
When precedent is thin, the agent reads chapter and section notes and applies the General Rules of Interpretation to pick the right heading.

      

      

        
3. Schedule verification

        
Every committed code is verified against the current 2026 HTS schedule. Older rulings often cite codes that have since been restructured — the agent finds the current equivalent.

      

      

        
4. Effective duty + evidence

        
Effective duty rates are pulled from US Census Bureau trade data — what importers actually paid. The response includes cited rulings, GRI notes, confidence, and a clarification question when the answer depends on info you didn't provide.

      

    

  







  

    
Data coverage and accuracy

    
HTS classification accuracy — agent-based v2 (2026)

    
Tested on 200 out-of-sample CBP rulings from 2024-2025. The agent searches ruling precedents, reads the USITC HTS schedule, and applies GRI rules.

    

      

        
70%
10-digit accuracy

      

      

        
70%
6-digit accuracy

      

      

        
80%
4-digit accuracy

      

    

    
Data sources updated for 2026

    

      
USITC Harmonized Tariff Schedule19,738 classifiable codes

      
CBP CROSS Customs Rulings134,050 rulings

      
US Census Bureau Trade DataReal-time effective rates

      
Section 301/232 tariff provisionsRate provision breakdown

      
GRI legal notesSection, Chapter, Heading

    

    
What makes this different from other HTS APIs

    
Most tariff classification APIs return a code and a rate. We return the code, the actual duty CBP collected from Census data (not a calculated estimate), the CBP rulings that support the classification, the official legal reasoning per WCO General Rules of Interpretation, and a specific question when the answer isn't clear. Every field is sourced and verifiable.

  







  

    
Frequently asked questions

    

      
How accurate is the HTS classification?

      
70% exact 10-digit accuracy and 80% at 4-digit heading on a 200-item benchmark of novel CBP rulings (products the agent hasn't seen before). The agent searches CBP precedent and applies GRI rules, telling you its confidence level and asking clarification questions when uncertain — so you know when to trust the result and when to verify.


      
Where do the duty rates come from?

      
Effective duty rates come from the US Census Bureau International Trade data — what CBP actually collected at the port. This includes MFN base rates, Section 301 and 232 tariffs, FTA program usage, and exclusions. Not a theoretical calculation.


      
Does the API cover Section 301 tariffs on China?

      
Yes. The rate provision breakdown shows exactly how much of the duty is from Section 301/232/reciprocal tariffs versus the MFN base rate. For example, a yoga mat from China shows 38.3% effective duty with 99% of imports hit by additional tariffs at the RP=69 level.


      
How does this compare to Zonos, Tarifflo, or Quickcode?

      
We're the only API that returns actual Census Bureau duty data alongside classification. Other APIs calculate duty from the tariff schedule — we show what was actually collected. We also return CBP ruling precedents and GRI legal notes, which no other classification API provides. Price: $0.05-$0.10 per classification.


      
Can I try it for free?

      
Yes. The web demo above is free with no API key required. Describe any product and get the full classification with duty rates, rulings, and legal notes.


      
What tariff schedule year does the API use?

      
The 2026 USITC Harmonized Tariff Schedule, refreshed on each HTS revision. Census Bureau duty data reflects the most recent available month (typically 6 weeks lag).

    

  







  

    
Pricing

    
No subscriptions. Credits never expire. Each /classify call = 1 credit.

    

      

        
100 Credits

        
$10

        
$0.10 per classification

        

          
Census effective duty rates

          
CBP ruling evidence

          
GRI legal notes

          
Progressive clarification

          
All other endpoints free

        

        
      

      

        
1,000 Credits

        
$50

        
$0.05 per classification

        

          
Everything above

          
50% cheaper per call

          
Priority support

        

        
      

    

    
Already have an API key? Sign in

    
Need volume pricing? Contact us.