# Klatchit

> Klatchit is a US marketplace where independent sellers list offers against a shared product catalog, and shoppers compare those offers side by side on one product page. Every listing is a competing offer on a canonical product: prices, conditions, and stores differ, the product does not. Payments go direct to the seller of the chosen offer. Prices are US dollars, quoted everywhere in integer minor units (cents).

## Product pages

Every catalog product has one canonical, human-readable page:

  https://www.klatchit.com/p/{slug}

The same `{slug}` addresses the product in the JSON API. Search result items
and API responses always carry the absolute URL, so you never need to build one
by hand.

## Machine-readable access

- OpenAPI contract: https://www.klatchit.com/api/public/v1/openapi.json
- Catalog search:   https://www.klatchit.com/api/public/v1/search?q=...
- Product detail:   https://www.klatchit.com/api/public/v1/products/{slug}
- Category tree:    https://www.klatchit.com/api/public/v1/categories
- Cart link (POST): https://www.klatchit.com/api/public/v1/cart-links
- Product feed (Google Shopping XML): https://www.klatchit.com/feeds/google-shopping

The API needs no key or account. Responses are `{ "data": ..., "meta": ... }`
JSON; errors are `{ "error": { "type", "code", "message", "request_id" } }`.
Start from the OpenAPI document — it is generated from the server's own
validation schemas, so it is never stale.

For the SAME contract as plain markdown — every endpoint with its parameters,
response shape, and error codes, plus the MCP and cart-link details — read:

  https://www.klatchit.com/llms-full.txt

## MCP

An MCP (Model Context Protocol) server is LIVE:

  https://www.klatchit.com/api/mcp/mcp

Streamable HTTP, `POST` only, stateless, no authentication. It exposes seven
tools: three read-only catalog tools — `search_products`, `get_product`,
`get_categories` — two read-only help tools — `search_help`, `get_help_article`
(the /help center: policies, fees, returns, shipping) — the `create_cart_link`
handoff (see Buying below), and `get_connection_guide` (how to connect a client).
Add it as a custom connector by URL; there is no auth step. Human documentation:
https://www.klatchit.com/developers/public

## Authenticated access

Everything above needs no account. A Klatchit account holder can also connect
THEIR OWN agent so it acts as them, within scopes they explicitly consent to:

- Authenticated MCP (OAuth): https://www.klatchit.com/api/agent/mcp — add it as a custom
  connector; the client completes an OAuth sign-in and consent, and the
  connection then reads that user's own orders, cart, and returns (or, for a
  store member, the store's order queue). Agentic checkout is possible there but
  gated by the marketplace and the buyer's consent + spending caps — never
  assume it is on.
- Seller REST API (store API keys): https://www.klatchit.com/developers
- Setup guides for Claude, ChatGPT, and your own code: https://www.klatchit.com/developers/connect

## What is exposed

Exactly what an anonymous visitor can see: publicly listed products, active
categories, public prices, and purchasable offers. Seller-internal data —
inventory counts, contact details, order and payout information — is not
available through any public endpoint.

## Buying

There is no way to buy through this API, by design. `POST /cart-links` mints a
link that a PERSON opens in their own browser to find the chosen offers waiting
in their own cart, which they then review and pay for themselves. It takes no
payment, places no order, and reserves no stock, and price and availability are
re-checked when the link is opened. Hand the link to your user; never report a
purchase you have not made.

## Usage

- Requests are rate limited per client IP. A `429` response carries a
  `Retry-After` header in seconds; honour it and back off.
- Send a descriptive `User-Agent` identifying your agent or crawler.
- Responses are cacheable (`Cache-Control: public, s-maxage=60`); please reuse
  them rather than re-fetching identical queries.
- https://www.klatchit.com/robots.txt governs crawling of the HTML site.
- Terms: https://www.klatchit.com/legal