How to integrate Semrush MCP with Vercel AI SDK

BOOK A CALLGET STARTED FOR FREE
Framework Integration Gradient
Semrush Logo
Vercel AI SDK Logo
divider

Introduction

This guide walks you through connecting Semrush to Vercel AI SDK using the Composio tool router. By the end, you'll have a working Semrush agent that can show top anchor texts for example.com, compare backlink profiles for three domains, get keyword overview for 'organic coffee', list ad copies seen for my competitor through natural language commands.

This guide will help you understand how to give your Vercel AI SDK agent real control over a Semrush account through Composio's Semrush MCP server.

Before we dive in, let's take a quick look at the key ideas and tools involved.

TL;DR

Here's what you'll learn:
  • How to set up and configure a Vercel AI SDK agent with Semrush integration
  • Using Composio's Tool Router to dynamically load and access Semrush tools
  • Creating an MCP client connection using HTTP transport
  • Building an interactive CLI chat interface with conversation history management
  • Handling tool calls and results within the Vercel AI SDK framework

What is Vercel AI SDK?

The Vercel AI SDK is a TypeScript library for building AI-powered applications. It provides tools for creating agents that can use external services and maintain conversation state.

Key features include:

  • streamText: Core function for streaming responses with real-time tool support
  • MCP Client: Built-in support for Model Context Protocol
  • Step Counting: Control multi-step tool execution
  • OpenAI Provider: Native integration with OpenAI models

What is the Semrush MCP server, and what's possible with it?

The Semrush MCP server is an implementation of the Model Context Protocol that connects your AI agent and assistants like Claude, Cursor, etc directly to your Semrush account. It provides structured and secure access to your SEO, keyword, and advertising analytics, so your agent can perform actions like keyword research, competitor analysis, backlink audits, and ad copy retrieval automatically on your behalf.

  • Comprehensive keyword research and reporting: Let your agent fetch broad match keywords, generate batch keyword overviews, and analyze key SEO metrics like search volume and difficulty in real time.
  • Competitor and backlink analysis: Ask your agent to pull backlink profiles, perform batch comparisons of domains, and summarize backlink authority and link types for competitive intelligence.
  • Ad campaign and copy insights: Have the agent retrieve unique Google Ads copies for any domain, helping you benchmark and optimize your own ad strategies based on real competitor data.
  • Content and category profiling: Enable your agent to analyze and categorize domains or URLs, surfacing topic strengths and audience focus areas for smarter content planning.
  • Anchor text and authority monitoring: Direct your agent to report on anchor text distributions and authority score profiles, giving you actionable insights for improving link-building efforts.

Supported Tools & Triggers

Tools
Get ad copiesRetrieves unique ad copies semrush has observed for a specified domain from a regional database, detailing ads seen in google's paid search results.
Get anchor textsUse this action to get a csv report of anchor texts for backlinks pointing to a specified, publicly accessible domain, root domain, or url.
Get authority score profileRetrieves the authority score (as) profile for a specified target, showing the count of referring domains that link to the target for each as value from 0 to 100.
Get backlinksFetches backlinks for a specified domain or url as a csv-formatted string, allowing customization of columns, sorting, and filtering; ensure `display limit` surpasses `display offset` when an offset is used, and note the `urlanchor` filter may have limitations for targets with extensive backlinks.
Backlinks overviewProvides a csv-formatted summary of backlinks, including authority score and link type breakdowns, for a specified and publicly accessible domain, root domain, or url.
Batch comparisonCompares backlink profiles for multiple specified targets (domains, subdomains, or urls) to analyze and compare link-building efforts.
Batch keyword overviewFetches a keyword overview report from a semrush regional database for up to 100 keywords, providing metrics like search volume, cpc, and keyword difficulty.
Broad match keywordFetches broad match keywords for a given phrase; `display sort` and `display filter` parameters are defined but currently not utilized by the api call.
Get categoriesRetrieves categories and their 0-1 confidence ratings for a specified domain, subdomain, or url, with results sorted by rating.
Get categories profileRetrieves a profile of content categories from referring domains for a specified target, analyzing its first 10,000 referring domains and sorting results by domain count.
Get competitor dataRetrieves a customizable csv report of competitors for a specified target (root domain, domain, or url) based on shared backlinks or referring domains, ensuring the target is valid and its type is correctly specified.
Get competitors in organic searchUse to get a domain's organic search competitors from semrush as a semicolon-separated string; `display date` requires 'yyyymm15' format if used.
Get competitors in paid searchRetrieves a list of a domain's competitors in paid search results from a specified regional database.
Get domain ad historyRetrieves a domain's 12-month advertising history from semrush (keywords bid on, ad positions, ad copy) for ppc strategy and competitor analysis; most effective when the domain has ad history in the selected database.
Get domain organic pagesFetches a report on a domain's unique organic pages ranking in google's top 100 search results, with options for specifying database, date, columns, sorting, and filtering.
Get domain organic search keywordsRetrieves organic search keywords for a domain from a specified semrush regional database; `display positions` must be set if `display daily=1` for daily updates.
Get domain organic subdomainsRetrieves a report on subdomains of a given domain that rank in google's top 100 organic search results for a specified regional database.
Get domain paid search keywordsFetches keywords driving paid search traffic to a specified, existing domain using a supported semrush regional database.
Get PLA search keywords for a domainRetrieves product listing ad (pla) search keywords for a specified domain from a semrush regional database.
Compare domainsAnalyzes keyword rankings by comparing up to five domains to find common, unique, or gap keywords, using specified organic/paid types and comparison logic in the `domains` string.
Get historical dataRetrieves monthly historical backlink and referring domain data for a specified root domain, returned as a time series string with newest records first.
Get indexed pagesRetrieves a list of indexed pages from semrush for a specified `target` (root domain, domain, or url) and `target type`, ensuring `target` is publicly accessible, semrush-analyzable, and correctly matches `target type`.
Get keyword difficultyDetermines the keyword difficulty (kd) score (0-100, higher means greater difficulty) for a given phrase in a specific semrush regional database to assess its seo competitiveness.
Keyword overview all databasesFetches a keyword overview from semrush for a specified phrase, including metrics like search volume, cpc, and competition.
Get keyword overview for one databaseFetches a keyword summary for a specified phrase from a chosen regional database.
Get keywords ads historyFetches a historical report (last 12 months) of domains advertising on a specified keyword in google ads, optionally for a specific month ('yyyymm15') or the most recent period, returning raw csv-like data.
Get organic resultsRetrieves up to 100,000 domains and urls from google's top 100 organic search results for a keyword and region, returning a raw string; use `display date` in 'yyyymm15' format (day must be '15') for historical data.
Get paid search resultsFetches domains ranking in google's paid search results (adwords) for a specified keyword and regional database.
Phrase questionsFetches question-format keywords semantically related to a given query phrase for a specified regional database, aiding in understanding user search intent and discovering content ideas.
Get PLA competitorsRetrieves domains competing with a specified domain in google's product listing ads (pla) from a given semrush regional database.
Get PLA copiesFetches product listing ad (pla) copies that semrush observed for a domain in google's paid search results.
Get referring domainsRetrieves a report as a text string (e.
Get referring domains by countryGenerates a csv report detailing the geographic distribution of referring domains (by country, determined via ip address) for a specified, publicly accessible target.
Referring i psFetches ip addresses that are sources of backlinks for a specified target domain, root domain, or url.
Find related keywordsCall this to find related keywords (including synonyms and variations) for a target phrase in a specific regional database; `display date` (if used for historical data) must be 'yyyymm15' for a past month.
Get TLD distributionFetches a report on the top-level domain (tld) distribution of referring domains for a specified target, useful for analyzing geographic or categorical backlink diversity.

What is the Composio tool router, and how does it fit here?

What is Tool Router?

Composio's Tool Router helps agents find the right tools for a task at runtime. You can plug in multiple toolkits (like Gmail, HubSpot, and GitHub), and the agent will identify the relevant app and action to complete multi-step workflows. This can reduce token usage and improve the reliability of tool calls. Read more here: Getting started with Tool Router

The tool router generates a secure MCP URL that your agents can access to perform actions.

How the Tool Router works

The Tool Router follows a three-phase workflow:

  1. Discovery: Searches for tools matching your task and returns relevant toolkits with their details.
  2. Authentication: Checks for active connections. If missing, creates an auth config and returns a connection URL via Auth Link.
  3. Execution: Executes the action using the authenticated connection.

Step-by-step Guide

Prerequisites

Before you begin, make sure you have:
  • Node.js and npm installed
  • A Composio account with API key
  • An OpenAI API key

Getting API Keys for OpenAI and Composio

OpenAI API Key
  • Go to the OpenAI dashboard and create an API key. You'll need credits to use the models, or you can connect to another model provider.
  • Keep the API key safe.
Composio API Key
  • Log in to the Composio dashboard.
  • Navigate to your API settings and generate a new API key.
  • Store this key securely as you'll need it for authentication.

Install required dependencies

bash
npm install @ai-sdk/openai @ai-sdk/mcp @composio/core ai dotenv

First, install the necessary packages for your project.

What you're installing:

  • @ai-sdk/openai: Vercel AI SDK's OpenAI provider
  • @ai-sdk/mcp: MCP client for Vercel AI SDK
  • @composio/core: Composio SDK for tool integration
  • ai: Core Vercel AI SDK
  • dotenv: Environment variable management

Set up environment variables

bash
OPENAI_API_KEY=your_openai_api_key_here
COMPOSIO_API_KEY=your_composio_api_key_here
COMPOSIO_USER_ID=your_user_id_here

Create a .env file in your project root.

What's needed:

  • OPENAI_API_KEY: Your OpenAI API key for GPT model access
  • COMPOSIO_API_KEY: Your Composio API key for tool access
  • COMPOSIO_USER_ID: A unique identifier for the user session

Import required modules and validate environment

typescript
import "dotenv/config";
import { openai } from "@ai-sdk/openai";
import { Composio } from "@composio/core";
import * as readline from "readline";
import { streamText, type ModelMessage, stepCountIs } from "ai";
import { experimental_createMCPClient as createMCPClient } from "@ai-sdk/mcp";

const composioAPIKey = process.env.COMPOSIO_API_KEY;
const composioUserID = process.env.COMPOSIO_USER_ID;

if (!process.env.OPENAI_API_KEY) throw new Error("OPENAI_API_KEY is not set");
if (!composioAPIKey) throw new Error("COMPOSIO_API_KEY is not set");
if (!composioUserID) throw new Error("COMPOSIO_USER_ID is not set");

const composio = new Composio({
  apiKey: composioAPIKey,
});
What's happening:
  • We're importing all necessary libraries including Vercel AI SDK's OpenAI provider and Composio
  • The dotenv/config import automatically loads environment variables
  • The MCP client import enables connection to Composio's tool server

Create Tool Router session and initialize MCP client

typescript
async function main() {
  // Create a tool router session for the user
  const { session } = await composio.create(composioUserID!, {
    toolkits: ["semrush"],
  });

  const mcpUrl = session.mcp.url;
What's happening:
  • We're creating a Tool Router session that gives your agent access to Semrush tools
  • The create method takes the user ID and specifies which toolkits should be available
  • The returned mcp object contains the URL and authentication headers needed to connect to the MCP server
  • This session provides access to all Semrush-related tools through the MCP protocol

Connect to MCP server and retrieve tools

typescript
const mcpClient = await createMCPClient({
  transport: {
    type: "http",
    url: mcpUrl,
    headers: session.mcp.headers, // Authentication headers for the Composio MCP server
  },
});

const tools = await mcpClient.tools();
What's happening:
  • We're creating an MCP client that connects to our Composio Tool Router session via HTTP
  • The mcp.url provides the endpoint, and mcp.headers contains authentication credentials
  • The type: "http" is important - Composio requires HTTP transport
  • tools() retrieves all available Semrush tools that the agent can use

Initialize conversation and CLI interface

typescript
let messages: ModelMessage[] = [];

console.log("Chat started! Type 'exit' or 'quit' to end the conversation.\n");
console.log(
  "Ask any questions related to semrush, like summarize my last 5 emails, send an email, etc... :)))\n",
);

const rl = readline.createInterface({
  input: process.stdin,
  output: process.stdout,
  prompt: "> ",
});

rl.prompt();
What's happening:
  • We initialize an empty messages array to maintain conversation history
  • A readline interface is created to accept user input from the command line
  • Instructions are displayed to guide the user on how to interact with the agent

Handle user input and stream responses with real-time tool feedback

typescript
rl.on("line", async (userInput: string) => {
  const trimmedInput = userInput.trim();

  if (["exit", "quit", "bye"].includes(trimmedInput.toLowerCase())) {
    console.log("\nGoodbye!");
    rl.close();
    process.exit(0);
  }

  if (!trimmedInput) {
    rl.prompt();
    return;
  }

  messages.push({ role: "user", content: trimmedInput });
  console.log("\nAgent is thinking...\n");

  try {
    const stream = streamText({
      model: openai("gpt-5"),
      messages,
      tools,
      toolChoice: "auto",
      stopWhen: stepCountIs(10),
      onStepFinish: (step) => {
        for (const toolCall of step.toolCalls) {
          console.log(`[Using tool: ${toolCall.toolName}]`);
          }
          if (step.toolCalls.length > 0) {
            console.log(""); // Add space after tool calls
          }
        },
      });

      for await (const chunk of stream.textStream) {
        process.stdout.write(chunk);
      }

      console.log("\n\n---\n");

      // Get final result for message history
      const response = await stream.response;
      if (response?.messages?.length) {
        messages.push(...response.messages);
      }
    } catch (error) {
      console.error("\nAn error occurred while talking to the agent:");
      console.error(error);
      console.log(
        "\nYou can try again or restart the app if it keeps happening.\n",
      );
    } finally {
      rl.prompt();
    }
  });

  rl.on("close", async () => {
    await mcpClient.close();
    console.log("\nđź‘‹ Session ended.");
    process.exit(0);
  });
}

main().catch((err) => {
  console.error("Fatal error:", err);
  process.exit(1);
});
What's happening:
  • We use streamText instead of generateText to stream responses in real-time
  • toolChoice: "auto" allows the model to decide when to use Semrush tools
  • stopWhen: stepCountIs(10) allows up to 10 steps for complex multi-tool operations
  • onStepFinish callback displays which tools are being used in real-time
  • We iterate through the text stream to create a typewriter effect as the agent responds
  • The complete response is added to conversation history to maintain context
  • Errors are caught and displayed with helpful retry suggestions

Complete Code

Here's the complete code to get you started with Semrush and Vercel AI SDK:

typescript
import "dotenv/config";
import { openai } from "@ai-sdk/openai";
import { Composio } from "@composio/core";
import * as readline from "readline";
import { streamText, type ModelMessage, stepCountIs } from "ai";
import { experimental_createMCPClient as createMCPClient } from "@ai-sdk/mcp";

const composioAPIKey = process.env.COMPOSIO_API_KEY;
const composioUserID = process.env.COMPOSIO_USER_ID;

if (!process.env.OPENAI_API_KEY) throw new Error("OPENAI_API_KEY is not set");
if (!composioAPIKey) throw new Error("COMPOSIO_API_KEY is not set");
if (!composioUserID) throw new Error("COMPOSIO_USER_ID is not set");

const composio = new Composio({
  apiKey: composioAPIKey,
});

async function main() {
  // Create a tool router session for the user
  const { session } = await composio.create(composioUserID!, {
    toolkits: ["semrush"],
  });

  const mcpUrl = session.mcp.url;

  const mcpClient = await createMCPClient({
    transport: {
      type: "http",
      url: mcpUrl,
      headers: session.mcp.headers, // Authentication headers for the Composio MCP server
    },
  });

  const tools = await mcpClient.tools();

  let messages: ModelMessage[] = [];

  console.log("Chat started! Type 'exit' or 'quit' to end the conversation.\n");
  console.log(
    "Ask any questions related to semrush, like summarize my last 5 emails, send an email, etc... :)))\n",
  );

  const rl = readline.createInterface({
    input: process.stdin,
    output: process.stdout,
    prompt: "> ",
  });

  rl.prompt();

  rl.on("line", async (userInput: string) => {
    const trimmedInput = userInput.trim();

    if (["exit", "quit", "bye"].includes(trimmedInput.toLowerCase())) {
      console.log("\nGoodbye!");
      rl.close();
      process.exit(0);
    }

    if (!trimmedInput) {
      rl.prompt();
      return;
    }

    messages.push({ role: "user", content: trimmedInput });
    console.log("\nAgent is thinking...\n");

    try {
      const stream = streamText({
        model: openai("gpt-5"),
        messages,
        tools,
        toolChoice: "auto",
        stopWhen: stepCountIs(10),
        onStepFinish: (step) => {
          for (const toolCall of step.toolCalls) {
            console.log(`[Using tool: ${toolCall.toolName}]`);
          }
          if (step.toolCalls.length > 0) {
            console.log(""); // Add space after tool calls
          }
        },
      });

      for await (const chunk of stream.textStream) {
        process.stdout.write(chunk);
      }

      console.log("\n\n---\n");

      // Get final result for message history
      const response = await stream.response;
      if (response?.messages?.length) {
        messages.push(...response.messages);
      }
    } catch (error) {
      console.error("\nAn error occurred while talking to the agent:");
      console.error(error);
      console.log(
        "\nYou can try again or restart the app if it keeps happening.\n",
      );
    } finally {
      rl.prompt();
    }
  });

  rl.on("close", async () => {
    await mcpClient.close();
    console.log("\nđź‘‹ Session ended.");
    process.exit(0);
  });
}

main().catch((err) => {
  console.error("Fatal error:", err);
  process.exit(1);
});

Conclusion

You've successfully built a Semrush agent using the Vercel AI SDK with streaming capabilities! This implementation provides a powerful foundation for building AI applications with natural language interfaces and real-time feedback.

Key features of this implementation:

  • Real-time streaming responses for a better user experience with typewriter effect
  • Live tool execution feedback showing which tools are being used as the agent works
  • Dynamic tool loading through Composio's Tool Router with secure authentication
  • Multi-step tool execution with configurable step limits (up to 10 steps)
  • Comprehensive error handling for robust agent execution
  • Conversation history maintenance for context-aware responses

You can extend this further by adding custom error handling, implementing specific business logic, or integrating additional Composio toolkits to create multi-app workflows.

How to build Semrush MCP Agent with another framework

OpenAI Agents SDK

OpenAI Agents SDK

Use Semrush MCP with OpenAI Agents SDK

Claude Agent SDK

Claude Agent SDK

Use Semrush MCP with Claude Agent SDK

Google ADK

Google ADK

Use Semrush MCP with Google ADK

LangChain

LangChain

Use Semrush MCP with LangChain

Mastra AI

Mastra AI

Use Semrush MCP with Mastra AI

LlamaIndex

LlamaIndex

Use Semrush MCP with LlamaIndex

CrewAI

CrewAI

Use Semrush MCP with CrewAI

FAQ

What are the differences in Tool Router MCP and Semrush MCP?

With a standalone Semrush MCP server, the agents and LLMs can only access a fixed set of Semrush tools tied to that server. However, with the Composio Tool Router, agents can dynamically load tools from Semrush and many other apps based on the task at hand, all through a single MCP endpoint.

Can I use Tool Router MCP with Vercel AI SDK?

Yes, you can. Vercel AI SDK fully supports MCP integration. You get structured tool calling, message history handling, and model orchestration while Tool Router takes care of discovering and serving the right Semrush tools.

Can I manage the permissions and scopes for Semrush while using Tool Router?

Yes, absolutely. You can configure which Semrush scopes and actions are allowed when connecting your account to Composio. You can also bring your own OAuth credentials or API configuration so you keep full control over what the agent can do.

How safe is my data with Composio Tool Router?

All sensitive data such as tokens, keys, and configuration is fully encrypted at rest and in transit. Composio is SOC 2 Type 2 compliant and follows strict security practices so your Semrush data and credentials are handled as safely as possible.

Used by agents from

Context
Letta
glean
HubSpot
Agent.ai
Altera
DataStax
Entelligence
Rolai
Context
Letta
glean
HubSpot
Agent.ai
Altera
DataStax
Entelligence
Rolai
Context
Letta
glean
HubSpot
Agent.ai
Altera
DataStax
Entelligence
Rolai

Never worry about agent reliability

We handle tool reliability, observability, and security so you never have to second-guess an agent action.

Book a call↗Get started for free