Search and read live eBay listings through the Buy Browse API.
eBay ships in the w6w first-party pack. It declares 4 actions, 2 health checks, and the host runs its code in a sandbox that never sees the credential.
io.w6w.ebayeBay covers the Buy Browse API — the public-listing surface of eBay’s REST platform, which lives on api.ebay.com alongside every other eBay REST API (Sell, Buy, Commerce). A keyword or category search returns a page of item summaries with price, condition, seller and shipping data; a single item (by its RESTful item ID, its legacy numeric ID, or as one variation in a multi-variation listing group) can be read in full.
Authentication is app-level rather than tied to any individual buyer or seller: eBay’s client-credentials grant mints an Application access token from a registered app’s Client ID and Secret, so this app works in scheduled and unattended runs with no browser sign-in step. That token authorizes only eBay’s public-data scope — it cannot see or touch any seller’s own inventory, orders or account, which is a different, user-consented part of eBay’s API surface this app does not cover.
A quota check reads eBay’s own Analytics API for real call-count headroom against the Browse API’s rate limit, using the same application token.
Three routes to the same 4 actions. The Workflow tab is generated from eBay's own manifest and carries its real ids, so it is copy-pasteable; the Code and CLI examples are the same call for any action on any app, so every app-specific value in them is a blank you fill in.
item-get-by-legacy-id Show full details for one item, by its legacy numeric eBay item ID.
item-group-get List every variation (size, color, ...) of a multi-SKU eBay listing.
A workflow step names the app and the action, and the editor fills in the
connection when you pick one. This is the Step shape from the
workflow spec, carrying eBay's real ids.
{
"manifestVersion": "2",
"name": "ebay-example",
"steps": [
{
"id": "item-get",
"uses": {
"app": "io.w6w.ebay",
"action": "item-get",
"connection": "conn_YOUR_CONNECTION_ID"
},
"with": {
"itemId": "<itemId>"
}
}
]
}item-get item-get-by-legacy-id item-group-get item-search Every app-specific value here is a blank you have to fill in. An
app action is reached through the connection that authenticates it, so the
address is a connection id, not the app id — and connections belong to your account,
so a public page cannot know yours. Create one for eBay, then fill in
the three blanks: conn_YOUR_CONNECTION_ID, the action key, and the
parameters that action declares. The call itself is real — the shape is transcribed
from the studio's own snippet builder, which prints the same kind of blanks — but
nothing in it is specific to eBay. The Workflow tab is where this app's
real ids are.
npm install @w6w/sdkyarn add @w6w/sdkpnpm add @w6w/sdkdeno add npm:@w6w/sdkimport { W6wClient, isActionRun } from "@w6w/sdk";
// Reads W6W_BASE_URL and W6W_TOKEN from the environment when omitted.
const client = new W6wClient();
const envelope = await client.run({
urn: "conn_YOUR_CONNECTION_ID",
action: "item-get",
payload: {
itemId: "<value>",
// marketplaceId: "<value>",
},
});
if (isActionRun(envelope)) console.log(envelope.value); npm install -g @w6w/cli w6w run conn_YOUR_CONNECTION_ID --action item-get --payload '{"itemId":"<value>"}' Give an AI agent eBay — without giving it eBay's credentials. One MCP endpoint exposes every app, function and workflow the caller is entitled to, as tools it can discover and run. Access is granted per team while we onboard.
One tool call{
"name": "w6w_invoke",
"arguments": {
"ref": "app:io.w6w.ebay#item-get",
"input": {
"itemId": "<itemId>"
}
}
}
Every tool names its target with a single ref. The
app: form above doesn't name a connection at all — the
host resolves which of the caller's eBay connections to sign
with, and refuses rather than guesses when the answer is ambiguous.
The token is attached host-side, at the moment of the call. It is never a tool argument, never in the model's context, and never in a transcript — so a prompt injection has nothing to exfiltrate.
Tools are derived per end user from what that person has actually connected and is entitled to — not one shared bot identity carrying the union of everyone's access.
Multi-step work runs on the workflow engine and returns a run handle the agent can poll — retries, branching and state survive the conversation that started them.
eBay's declared health checks are on the surface too, so an agent can tell "the vendor is down" from "your credential expired" before it burns a retry on either.
The MCP surface is part of the hosted platform. eBay itself is MIT, and the runtime that executes it is source-available (FSL).
eBay declares its own checks, so its health is a property of the app rather than something the host guesses at.
Per-resource call count, limit and reset time for the Browse API, read from the Developer Analytics API's GET /developer/analytics/v1_beta/rate_limit/.