Read SEMrush backlink, referring-domain, anchor, competitor and keyword reports, and check the account's remaining API units.
SEMrush ships in the w6w first-party pack. It declares 13 actions, 1 health check, and the host runs its code in a sandbox that never sees the credential.
io.w6w.semrushSEMrush is an SEO and marketing-analytics platform. This app reads its current v4 Standard API (api.semrush.com/apis/v4): a domain’s backlink profile — totals, historical summaries, individual links, referring domains and IPs, linked pages, anchors, the domain-score distribution and competitors — plus a keyword’s volume, difficulty and SERP features.
SEMrush meters these calls in paid ‘API units’, so the app also reads the account’s remaining balance from the one endpoint that costs nothing (www.semrush.com/users/countapiunits.html). That endpoint is also the credential probe, and it has a trap the app is built around: its error text echoes the submitted key back, so no vendor error string from it is ever surfaced.
Three routes to the same 13 actions. The Workflow tab is generated from SEMrush'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.
api-units-balance-get Read the account's remaining API units. This call itself costs none.
backlinks-anchors-get List the anchor texts pointing at a target, with link and domain counts.
backlinks-competitors-get List the domains with referring profiles most similar to a target domain.
backlinks-historical-summary-get Read a target's month-by-month backlink totals, oldest bucket included by date.
backlinks-list-get List the individual backlinks pointing at a target, with sorting and paging.
backlinks-matrix-get List referring domains with a per-target link count, for up to 6 targets.
backlinks-overview-get Read a target's backlink profile totals: links, domains, IPs and score.
backlinks-pages-get List the target's own pages that receive backlinks, with per-page link counts.
backlinks-referring-domains-get List the domains linking to a target, with per-domain counts and first-seen dates.
backlinks-referring-ips-get List the IP addresses linking to a target, with domain counts and locations.
backlinks-score-profile-get Read the distribution of referring domains across the 0-100 domain score range.
backlinks-summary-comparison-get Read month-by-month backlink totals for up to 10 targets in one call.
keyword-metrics-get Read a keyword's search volume, difficulty, CPC, intent and SERP features.
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 SEMrush's real ids.
{
"manifestVersion": "2",
"name": "semrush-example",
"steps": [
{
"id": "backlinks-anchors-get",
"uses": {
"app": "io.w6w.semrush",
"action": "backlinks-anchors-get",
"connection": "conn_YOUR_CONNECTION_ID"
},
"with": {
"url": "<url>",
"scope": "<scope>"
}
}
]
}backlinks-anchors-get api-units-balance-get backlinks-competitors-get backlinks-historical-summary-get backlinks-list-get +8 more actions available
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 SEMrush, 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 SEMrush. 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: "backlinks-anchors-get",
payload: {
url: "<value>",
scope: "<value>",
// fields: "<value>",
// order_by: "<value>",
// direction: "<value>",
// limit: "<value>",
// offset: "<value>",
// filter: "<value>",
},
});
if (isActionRun(envelope)) console.log(envelope.value); npm install -g @w6w/cli w6w run conn_YOUR_CONNECTION_ID --action backlinks-anchors-get --payload '{"url":"<value>","scope":"<value>"}' Give an AI agent SEMrush — without giving it SEMrush'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.semrush#backlinks-anchors-get",
"input": {
"url": "<url>",
"scope": "<scope>"
}
}
}
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 SEMrush 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.
SEMrush'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. SEMrush itself is MIT, and the runtime that executes it is source-available (FSL).
SEMrush declares its own checks, so its health is a property of the app rather than something the host guesses at.
Unauthenticated GET /apis/v4/backlinks/v1/overview against api.semrush.com. A schema-correct 401 error envelope is a pass: it proves the API parsed the request and answered with its own documented error shape. Whether a given key is live is the derived auth check's job.