Semrush MCP: Official Server, Setup, and Limits
Yes — Semrush has an official MCP server. Semrush’s remote server (mcp.semrush.com) lets any MCP-compatible AI tool pull live keyword, traffic, competitive, and backlink data straight from your Semrush account. It’s listed as an official connector in Claude, ChatGPT, and Perplexity, so if you’re searching “Semrush MCP,” the connection you want really does exist.
The thing worth knowing before you set it up: an MCP server hands your Semrush data to an AI inside a conversation you start. It’s a doorway, not a worker. Nothing watches your rankings for you, nothing fires when a keyword drops out of the top 10, and nothing runs while the chat is closed. Here’s exactly what the Semrush MCP does, how to turn it on, where it stops — and what to use when you want Semrush work that runs on its own.
For the wider picture, every app with an MCP server is listed by category.
What the Semrush MCP server does
Model Context Protocol (MCP) is the open standard that lets an AI client — Claude, ChatGPT, Cursor, VS Code, and others — talk to an outside app through a shared interface. Semrush’s official server, live since September 2025, connects two of its APIs:
- Standard API — core SEO data: keyword research, backlink research, organic and paid research, and domain analytics.
- Trends API — market and competitive intelligence: traffic estimates, audience data, and growth benchmarks for any domain.
- Projects API (read-only) — pulls data from your existing Semrush projects without letting the AI create or change project settings.
With it connected, an AI client can:
- Pull keyword data on demand — “what’s the search volume and difficulty for ‘ai personal assistant’?” answered from live Semrush data, not a guess.
- Check rankings and competitors — “which keywords does this competitor rank for that we don’t?” run against real overlap data.
- Pull traffic and backlink numbers — estimate a domain’s traffic mix or audit its backlink profile mid-conversation.
- Reason across reports — summarize a keyword gap or a traffic trend without you exporting a CSV first.
It’s genuinely useful for ad-hoc research: ask a question, get an answer grounded in real Semrush data, without switching tabs.
How to set up the Semrush MCP server
The remote server is the quick path — no code, no hosting:
- In your AI client’s connector settings, add the Semrush MCP endpoint (
https://mcp.semrush.com/v2/mcp), or install it as an official app if your client offers Semrush directly (Claude and ChatGPT both do). - Authenticate — either OAuth through your Semrush login, or an API key passed in the Authorization header.
- Confirm the tools appear in the client, then start a chat and ask it to pull a keyword, traffic, or backlink report.
Access requires a Semrush plan with API units — Semrush One Starter and up, or SEO Classic Pro and up, each with an included unit allowance you can top up. There’s nothing to self-host; the server is entirely Semrush’s.
Where the Semrush MCP stops
None of this is a knock on MCP — it’s just the shape of the protocol. Four limits show up the moment you want more than a conversation:
- It only works inside a chat you start. Close the window and nothing happens. The AI doesn’t watch your rankings; it waits for you to ask.
- No triggers. A keyword falling out of the top 10, a competitor publishing new content, a sudden traffic dip — none of these can start anything through MCP. There’s no “when this happens in Semrush, do that.”
- It’s one app at a time. The Semrush MCP knows Semrush. Getting a ranking drop into a Slack alert, a client report, and a task on your project board means wiring up a separate MCP server for each and hoping your client can juggle them in one turn.
- You own the plumbing and the scopes. The API units, OAuth tokens, and how much account access you’re granting the AI are all on you to track.
- Read-only, mostly. The Semrush MCP is built for pulling data, not managing it — it won’t create Semrush projects or change settings on your behalf.
So the Semrush MCP is a great way to ask Semrush things mid-chat. It is not a way to make SEO monitoring run — to have work happen on a schedule or in reaction to a ranking or traffic change, across the other tools that report depends on.
Running Semrush work that doesn’t need a chat open
That “run on its own, across apps” gap is exactly where Carly fits. Carly connects to Semrush data through your own API key — no MCP endpoint to wire up, no client-by-client setup — and to the ~260 other apps it supports natively, plus anything else with a public API. The difference from MCP is the important part: Carly’s workflows are triggered and scheduled, so Semrush-driven work happens whether or not anyone has a chat window open.
A few things that MCP can’t do but a Carly workflow can:
- Every Monday morning → pull last week’s keyword rankings, summarize the biggest gains and drops, and email the list to the team.
- When a tracked keyword drops out of the top 10 → post an alert to Slack and create a follow-up task for the SEO owner.
- Once a month → pull a competitor’s new backlinks and traffic trend, draft a summary, and drop it into a shared doc for review.
The non-AI steps — pulling the report, matching it against last week’s numbers, routing the result to Slack or a doc — are free and unlimited, the Zapier-style backbone of the workflow. The AI steps (summarizing, deciding what’s worth flagging) start at $35/month. You describe the outcome in plain language and Carly wires up the Semrush data and everything downstream.
If you just want to ask Semrush questions from a chat, its official MCP server is the right tool and it’s included with your plan. If you want SEO monitoring to actually run — on a trigger, on a schedule, across every app a ranking change touches — that’s the job MCP wasn’t built for, and it’s the one Carly was.
FAQ
Does Semrush have an official MCP server? Yes. Semrush’s remote MCP server, live since September 2025, gives MCP-compatible AI tools access to its Standard API (keywords, backlinks, domain analytics) and Trends API (traffic and competitive data), plus read-only access to your Projects.
Is the Semrush MCP server free? Connecting it is free, but using it isn’t — it draws from the API units included in your Semrush plan (Semrush One Starter and up, or SEO Classic Pro and up). Heavy use may require buying additional API unit packages.
Can the Semrush MCP trigger automations? No. MCP is request/response inside an AI chat — it has no triggers and nothing runs when the conversation is closed. For alerts when a ranking drops or a scheduled weekly report, you need a workflow tool like Carly rather than an MCP server.
What AI tools can connect to Semrush over MCP? Any MCP-compatible client can connect to the endpoint directly, and it’s also listed as an official app inside Claude, ChatGPT, and Perplexity.
Can I connect Semrush to AI without coding or hosting a server? Yes. You don’t have to touch MCP at all. Carly connects to your Semrush data using your own API key and lets you build the automation in plain language — describe what you want to happen and it wires up Semrush and the other apps involved, with no server to host and no code to write.
Ready to automate your busywork?
Carly schedules, researches, and briefs you—so you can focus on what matters.
See what people say
"Before Carly, I relied on a Calendly link, but the whole process felt impersonal and not very professional. Carly changed that by handling all the back-and-forth, so I'm no longer stuck in endless email threads trying to line up schedules.
Now Carly reaches out to candidates, shares my real-time availability, lets them pick a slot, then sends a Zoom link and drops it straight into my calendar. She sends reminders to both of us before each call, which has significantly reduced no-shows and last-minute confusion.
On top of scheduling, Carly acts like a full executive assistant, sending me my schedule the night before so I can prepare for each call. It reminds me of the old x.ai assistant, but Carly is noticeably smarter, faster, and better suited to my healthcare recruitment business."
