Agentic Commerce

    The Data Layer for AI Shopping Agents

    Your agent is only as accurate as the data behind it. kwery returns GTIN-matched, confidence-scored prices from 12 marketplaces in one tool call.

    Get API KeyDocs
    · Free tier available · Plans from €79/mo

    12 Marketplaces · Live Offers with Trust Scores

    The Gap

    What Merchant Agents Can't See

    Merchant agents can read a retailer's own sales, stock, and orders, and draft price and campaign changes from that. What they can't see is the market outside the store, what a competitor charges right now, or whether a listed price is still current. kwery is the tool call that closes that gap, five things every price-data result needs, that most feeds don't provide:

    1. Verified identity

      GTIN-matched, not keyword-guessed, so the agent has the actual product, not something similar.

    2. A confidence score per result

      Identity, completeness, price integrity, offer contamination, and freshness, scored individually, so an agent can weigh how much to trust an answer instead of treating every result as equally reliable.

    3. A live, on-demand result

      Fetched at the moment of the call, not read from yesterday's crawl.

    4. All competing offers

      Every competing offer for that product in one result, merchant, price, availability, and trust, not a single price, so the agent compares the full market, not one data point.

    5. One schema, every source

      One structured JSON schema across every source, no HTML to parse, no per-marketplace normalization for the agent to maintain.

    View API Reference
    POST · api.kwery.co/job
    // One call per source — same GTIN, three markets compared
// POST /job {"source":"idealo","country":"de","key":"gtin","values":["4006381333962"]}
{ "name": "STABILO point 88 ocker", "price_min": 0.63, "offers_count": 13, "trust": { "score": 0.94 } }

// POST /job {"source":"ebay","country":"de","key":"gtin","values":["4006381333962"]}
{ "price_min": 1.00, "offers_count": 24, "trust": { "score": 0.83 } }

// POST /job {"source":"guenstiger","country":"de","key":"gtin","topic":"search_and_offers","values":["4006381333962"]}
{ "name": "Stabilo Point 88", "price_min": 0.64, "offers_count": 200, "trust": { "score": 0.80 } }
    See the Full Response

    Marketplace Coverage

    One Tool Call, 12 Marketplaces

    An agent can't predict which product a shopper will ask about next. kwery covers Amazon, Google Shopping, eBay, Walmart, Idealo, Otto, Kaufland, billiger.de, Guenstiger, Lidl, AliExpress, and Rakuten, one schema, one credential, no separate integration per source.

    View Source Library
    Amazon
    Global Coverage

    Amazon

    Pricing, Buy Box, offers, ratings, and seller data across 22 Amazon marketplaces.

    View API details →
    Google Shopping
    Global Coverage

    Google Shopping

    Live comparison prices and merchant listings across 48 supported markets.

    View API details →
    ebay
    Global Coverage

    ebay

    Offers, sold listings, bidding data, and seller analytics from 18 ebay markets.

    View API details →
    Walmart
    North America & Mexico

    Walmart

    Prices, availability, and seller data from Walmart US, Canada and Mexico.

    View API details →
    Idealo
    Europe

    Idealo

    Full merchant offers, price comparisons, and ranking data across DACH and Europe.

    View API details →
    Otto
    Germany

    Otto

    Product pricing and marketplace data from Germany's largest non-Amazon retailer.

    View API details →

    Try It

    Ready to point an agent at live market data?

    Start free with 1.000 credits. No credit card. One endpoint, every source.

    Get API KeyView Pricing

    Why kwery

    Built for One Request,
    One Answer

    Most price-data APIs are built for bulk jobs, submit thousands of SKUs, poll, download later. An agent making a decision mid-conversation can't wait on a queue. kwery's wait parameter holds the request open and returns the result of a single lookup in the same response. It's usually no polling: if a lookup takes longer than the wait limit, you get a job ID to poll instead.

    Every result is resolved live from the marketplace at the moment of the request, never from a stale or aggregated feed. You control exactly how fresh you need it with the max_age parameter: set it to 0 for a guaranteed live fetch every time, or allow a window so a request doesn't trigger a redundant fetch. One call in. One matched product, with its confidence score, back before the agent's next step.

    POST · api.kwery.co/job
    curl -s \
  https://api.kwery.co/job \
  -H "Authorization: Bearer $KWERY_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "source": "amazon",
    "country": "de",
    "key": "gtin",
    "values": ["4006381333962"],
    "wait": true
  }'

    Early Use Cases

    Who's Building With This

    Three ways teams are wiring in AI-ready price data.

    01

    Retailers building their own shopping assistant

    Feed live competitor and catalog pricing into a customer-facing agent without maintaining a separate data pipeline.

    Feeds → customer-facing agent

    02

    Marketplaces and comparison tools building AI-driven search

    GTIN matching means the agent compares the same product across sources, not similarly-titled listings.

    Feeds → AI search index

    03

    Teams building merchant agents for pricing and promotions

    A staff-facing agent that drafts pricing and promotion changes needs to know what the market is doing, not just what's in the retailer's own sales data.

    Feeds → merchant agent

    AI Integration

    Integrating With Claude Specifically?

    See how live marketplace data fits into a Claude merchant-agent workflow.

    See the Integration Guide

    No Signup Needed

    Try It Before You Build

    kwery ships a mock server that returns real response shapes with no API key and no credits consumed, prototype a tool definition, test how your agent handles the response, then swap in a live key when you're ready for production.

    API Reference

    POST · docs.kwery.co/_mock
    curl -X POST https://docs.kwery.co/_mock/api-reference/job \
  -H "Authorization: Bearer test123" \
  -H "Content-Type: application/json" \
  -H "x-redocly-response-body-example: wait" \
  -d '{"source":"idealo","country":"de","key":"gtin","values":["4006381333962"]}'

    Self-Serve

    One Credit. One Resolved SKU

    Product matching, product page, and offers, billed as a single credit.

    01

    Product Matching

    Identify the correct product within the requested source using GTIN or URL.

    02

    Product Page

    Title, brand, specs, images, and category, fully normalized.

    03

    Offers

    Every seller, price, condition, and shipping option in the offer section.

    Entry Price

    €79 / mo

    Included

    10k credits / mo

    Included Usage

    Up to 20k SKUs / mo

    Amazon, Walmart & ebay count 2 SKUs per credit. Every other source counts 1.

    Get API KeyView Pricing

    Free tier available · 1.000 one-time credits, no credit card required.

    Enterprise

    Need Custom Pricing, SLAs, or Support?

    Dedicated volume pricing, custom SLAs, and direct engineering support for teams running millions of SKUs.

    Talk to Sales

    Support

    Common Questions

    What teams ask before wiring kwery into an agent's tool-calling layer.

    Browse all FAQs

    Not as a packaged connector, and that's by design, not a gap. kwery is a plain REST/JSON API, so it already works from Claude, ChatGPT, LangChain, or any agent framework that can call an HTTP endpoint. A dedicated MCP connector would lock the integration into one specific path; staying a plain API keeps it usable everywhere without a separate wrapper to maintain.

    kwery isn't an official Anthropic partner on this, but it fits naturally: the blueprint's shopping agent expects a product-search/catalog tool behind it, and kwery's API is a straightforward fit to wire in as that tool's data source for cross-marketplace price and offer lookups, you connect it yourself using the blueprint's existing tool-contract pattern.

    For a single lookup, yes: set wait and the result comes back in the same response (up to 120 s). If a lookup takes longer than the wait limit, you get a job ID to poll instead. Batches run as jobs. Webhooks exist separately for streaming/monitoring use cases, not because single lookups need them.

    A feed hands you whatever was last scraped on someone else's schedule. kwery resolves live on each request, cached results are only served within a freshness window you set yourself (up to 3 days), so a call outside that window always triggers a real fetch, not stale data with no way to know how old it is.

    Yes, it's a standard REST/JSON API underneath, so it's callable from any framework that can make an HTTP request: LangChain tools, OpenAI function calling, or a custom agent loop, not just Claude.

    Yes, kwery ships a mock server that returns real response shapes with no API key and no credits consumed. Prototype a tool definition, test how your agent handles the response, then swap in a live key when you're ready for production.

    Give Your Agent Data It Can Trust

    GTIN-matched, confidence-scored pricing across 12 marketplaces, one tool call, one answer.

    Get API KeyView Documentation
    Enterprise SLA
    Global Scale