All posts
Guide

Use SearchFit Directly From Claude Desktop

Connect SearchFit to Claude Desktop in 5 minutes. Track AI visibility for ChatGPT, Perplexity, Gemini directly from Claude with MCP server setup.

KKevin12 minutes read

Use SearchFit Directly From Claude Desktop

Why Connect SearchFit to Claude Desktop?

Your competitors are already getting cited in ChatGPT, Perplexity, and Gemini. Meanwhile, you're juggling multiple tabs, flipping between SearchFit's dashboard, Claude, and your content calendar. That workflow costs time and introduces friction into your answer engine optimization process.

Connecting SearchFit's MCP server to Claude Desktop eliminates that friction. You can now query your AI visibility metrics, analyze citation patterns, and generate AEO-optimized content without leaving Claude. No more context switching. No more manual data pulls. Just direct, real-time access to your brand's performance across ChatGPT, Perplexity, Gemini, Claude, Grok, and Google.

This integration is built on Claude's Model Context Protocol (MCP), a standard that lets you extend Claude's capabilities with external tools. Once configured, you'll be able to ask Claude questions like "What's my citation rate in Perplexity this week?" or "Generate a blog outline optimized for Gemini's ranking factors," and get answers backed by live SearchFit data.

The setup takes about five minutes. Here's exactly how to do it.

Prerequisites: What You Need Before Starting

Before you connect SearchFit to Claude Desktop, confirm you have these items in place:

Claude Desktop Application You must be running Claude Desktop (the native application, not the web version). Download the latest version from Anthropic's official site. The MCP integration requires Claude 3.5 Sonnet or later, which includes native computer use and agentic capabilities as detailed in Anthropic's Claude 3.5 Sonnet announcement.

SearchFit Account with API Access You need an active SearchFit account with API credentials enabled. If you haven't signed up yet, create an account and navigate to your settings to generate an API key. This key authenticates the MCP server and lets Claude pull your AI tracking data directly.

API Key from SearchFit Locate your SearchFit API key in your account dashboard. You'll paste this into your Claude configuration file in the next step. Keep this key secure, treat it like a password.

Text Editor or Terminal Access You'll need to edit a JSON configuration file. Use any text editor (VS Code, Sublime, Notepad++) or your system's terminal. Mac users can use Terminal; Windows users can use PowerShell or Command Prompt.

Stable Internet Connection The MCP server communicates with SearchFit's backend in real time. A stable internet connection ensures smooth data retrieval.

Step 1: Locate Your Claude Desktop Configuration File

Claude Desktop stores its MCP server configuration in a specific location on your computer. Finding this file is the first step.

On macOS: Open Terminal and navigate to your Claude config directory:

~/.config/Claude/claude_desktop_config.json

If this directory doesn't exist, you'll create it in the next step.

On Windows: Open PowerShell or Command Prompt and navigate to:

%APPDATA%\Claude\claude_desktop_config.json

On Windows, %APPDATA% typically resolves to C:\Users\[YourUsername]\AppData\Roaming.

On Linux: The config file lives at:

~/.config/Claude/claude_desktop_config.json

If you're unsure of your system, open your file manager and search for claude_desktop_config.json. If the file doesn't exist yet, you'll create it in the next step. This is normal, Claude creates the configuration file once you've added your first MCP server.

Step 2: Create or Edit Your Configuration File

Now you'll add the SearchFit MCP server to your Claude configuration. Open the claude_desktop_config.json file in your text editor.

If the file is empty or doesn't exist, start with this template:

{
"mcpServers": {
"searchfit": {
"command": "node",
"args": ["/path/to/searchfit-mcp-server.js"],
"env": {
"SEARCHFIT_API_KEY": "your-api-key-here"
}
}
}
}

If the file already contains other MCP servers, add the SearchFit configuration inside the mcpServers object. For example:

{
"mcpServers": {
"existing-server": {
"command": "node",
"args": ["/path/to/existing-server.js"]
},
"searchfit": {
"command": "node",
"args": ["/path/to/searchfit-mcp-server.js"],
"env": {
"SEARCHFIT_API_KEY": "your-api-key-here"
}
}
}
}

Replace your-api-key-here with your actual SearchFit API key. Keep the quotes around the key, JSON format requires string values.

Replace /path/to/searchfit-mcp-server.js with the actual path to the SearchFit MCP server file on your system. If you've installed SearchFit's MCP package globally via npm, the path might look like /usr/local/lib/node_modules/searchfit-mcp-server/index.js (macOS/Linux) or C:\Users\[YourUsername]\AppData\Roaming\npm\node_modules\searchfit-mcp-server\index.js (Windows).

Step 3: Install the SearchFit MCP Server Package

If you haven't already installed the SearchFit MCP server locally, you'll need to do that now. The easiest way is through npm (Node Package Manager).

Install Node.js (if you don't have it) If you don't have Node.js installed, download it from nodejs.org. The LTS (Long Term Support) version is recommended for stability.

Verify your installation by opening Terminal or PowerShell and typing:

node --version

You should see a version number like v18.0.0 or higher.

Install SearchFit MCP Server Open Terminal (macOS/Linux) or PowerShell (Windows) and run:

npm install -g searchfit-mcp-server

The -g flag installs the package globally, making it accessible from anywhere on your system. This command downloads the SearchFit MCP server package and sets it up on your machine.

Wait for the installation to complete. You'll see output confirming the package version and location.

Verify the Installation After installation, confirm the server is installed by running:

npm list -g searchfit-mcp-server

You should see the package listed with its version number. If you see an error, try reinstalling or check your npm permissions.

Step 4: Update Your Configuration File with the Correct Path

Now that you've installed the SearchFit MCP server, update your claude_desktop_config.json file with the correct path to the installed package.

Find the Exact Installation Path Run this command to find where npm installed the package:

npm list -g searchfit-mcp-server --long

Look for the path in the output. On macOS/Linux, it typically looks like:

/usr/local/lib/node_modules/searchfit-mcp-server

On Windows, it might be:

C:\Users\[YourUsername]\AppData\Roaming\npm\node_modules\searchfit-mcp-server

Update the Configuration Open claude_desktop_config.json again and update the args path. If the installation path is /usr/local/lib/node_modules/searchfit-mcp-server, your config should look like:

{
"mcpServers": {
"searchfit": {
"command": "node",
"args": ["/usr/local/lib/node_modules/searchfit-mcp-server/index.js"],
"env": {
"SEARCHFIT_API_KEY": "your-actual-api-key"
}
}
}
}

Make sure your API key is in place and the path is correct. Save the file.

Step 5: Restart Claude Desktop

Configuration changes take effect only after you restart Claude. Close the Claude Desktop application completely, don't just minimize it.

On macOS: Click the Claude icon in the menu bar, then select "Quit Claude." Or use Command+Q.

On Windows: Click the X button to close Claude. Or use Alt+F4.

On Linux: Close the window or use the standard close command for your desktop environment.

Wait a few seconds, then reopen Claude Desktop. The application will load with your new MCP server configuration.

Step 6: Test the Connection

Once Claude has restarted, test whether the SearchFit MCP server is connected and working.

Open a new chat in Claude and type a simple query:

What's my current AI visibility across ChatGPT, Perplexity, and Gemini?

If the connection is successful, Claude will use the SearchFit MCP server to fetch your AI tracking data and return a response with your current metrics.

If you see an error message, double-check the following:

  • API Key: Verify your SearchFit API key is correct and hasn't expired. Regenerate it in your SearchFit account if needed.
  • File Path: Confirm the path to the SearchFit MCP server file is correct. Use the full path from Step 4.
  • JSON Syntax: Ensure your claude_desktop_config.json file has valid JSON syntax. Use an online JSON validator if you're unsure.
  • Claude Version: Make sure you're running Claude 3.5 Sonnet or later. Older versions don't support MCP servers.
  • Node.js: Verify Node.js is installed and accessible from your terminal.

If you're still having issues, check the Claude Desktop logs. On macOS, logs are typically in ~/Library/Logs/Claude. On Windows, check %APPDATA%\Claude\logs.

Using SearchFit Data in Claude: Practical Examples

Once your connection is working, you can leverage SearchFit data in Claude for real AEO work. Here are concrete use cases:

Query Your Citation Metrics Ask Claude for your citation breakdown:

Show me my top cited pages in Perplexity for the last 30 days. Which topics are driving the most citations?

Claude will pull this from your SearchFit analytics dashboard and present it in a readable format.

Analyze Competitor Visibility Compare your visibility to competitors:

Compare my ChatGPT citation rate to [competitor domain]. Where are they getting cited that I'm not?

This helps you identify content gaps and opportunities.

Generate AEO-Optimized Content Use SearchFit data to inform content creation:

Based on my SearchFit data, what topics should I cover in a new blog post to improve my Gemini visibility? Generate an outline optimized for AI search.

Claude combines your actual performance data with its knowledge of ranking factors to create targeted outlines.

Track Performance Over Time Monitor trends:

Has my Claude visibility improved in the last month? Show me the trend.

Claude can pull historical data and visualize the trend for you.

Audit Your AI Content Strategy Get strategic recommendations:

Based on my SearchFit metrics across all platforms, what's my biggest opportunity to improve AI visibility in the next quarter?

This combines your data with strategic thinking to guide your AEO roadmap.

Pro Tips for Maximizing the Integration

Batch Your Queries Instead of asking Claude one question at a time, batch multiple related questions into a single prompt. This reduces latency and gives Claude more context to work with. For example:

Give me a full AI visibility audit: (1) My top 5 cited pages across all platforms, (2) My citation rate trend for the last 90 days, (3) Comparison to my top 3 competitors, and (4) Top 3 recommendations to improve Perplexity visibility.

Combine SearchFit Data with Web Search Claude can now integrate SearchFit data with live web search. As detailed in Claude's web search documentation, you can ask Claude to search the web for competitor content while referencing your SearchFit metrics. This gives you a complete competitive intelligence loop.

Export Insights for Stakeholders Claude can format SearchFit data into reports, slide decks, or email summaries. Ask:

Create an executive summary of my AI visibility for Q4, formatted as a 1-page memo for my CMO.

Claude will pull your SearchFit data and present it in a stakeholder-friendly format.

Automate Content Audits Use Claude to systematically audit your content against AEO best practices. Reference your SearchFit data to identify which pages are underperforming in AI search and why.

Set Up Recurring Briefings Create a prompt template you can reuse weekly or monthly:

[Your recurring briefing template]
What changed in my AI visibility this week? Show me gains and losses by platform, and recommend 2 quick wins I can implement.

Save this prompt and run it on a schedule to stay on top of trends.

Troubleshooting Common Issues

"MCP Server Not Found" Error This usually means Claude can't locate the SearchFit MCP server file. Solutions:

  • Verify the file path in your config is absolute (starts with / on macOS/Linux or C:\ on Windows), not relative.
  • Confirm the file exists at that path by opening it in your file manager.
  • Reinstall the SearchFit MCP server package with npm install -g searchfit-mcp-server.
  • Restart Claude after making changes.

"API Key Invalid" Error Your SearchFit API key isn't being recognized. Solutions:

  • Log into your SearchFit account and verify your API key hasn't expired.
  • Regenerate a new API key in your account settings.
  • Ensure there are no extra spaces or special characters around your API key in the config file.
  • Double-check that you're using the correct API key (not your password).

Slow Response Times If Claude is slow to return SearchFit data, consider:

  • Checking your internet connection speed.
  • Reducing the scope of your query (e.g., ask for the last 7 days instead of 90).
  • Checking SearchFit's status page to ensure their API is running normally.
  • Closing other applications consuming bandwidth.

Configuration File Won't Save If you can't save your claude_desktop_config.json file:

  • Ensure you have write permissions for the directory.
  • On macOS/Linux, you may need to use sudo to edit files in .config directories.
  • Try saving to a different location first to confirm your text editor is working.
  • Use terminal commands (nano, vim) instead of a GUI text editor if permissions are the issue.

Claude Doesn't Recognize SearchFit Commands If Claude isn't using the SearchFit MCP server even after restart:

  • Verify the claude_desktop_config.json file has valid JSON syntax (use a JSON validator).
  • Confirm Node.js is installed and the node command works in your terminal.
  • Check that the SearchFit MCP server package is actually installed with npm list -g searchfit-mcp-server.
  • Look at Claude's logs for error messages that might reveal the issue.
  • Try restarting your computer (not just Claude) to ensure all system processes reload.

Understanding MCP and What It Enables

The Model Context Protocol (MCP) is a standard that lets Claude interact with external tools and data sources. By connecting SearchFit via MCP, you're essentially giving Claude direct access to your AI visibility data.

This is different from simply copying and pasting data into Claude. With MCP, Claude can:

  • Query SearchFit data in real time (not cached or stale)
  • Integrate SearchFit insights with other tools and web search
  • Combine your actual performance data with strategic analysis
  • Automate repetitive tasks like weekly audits or competitor tracking

For more context on how MCP works with Claude's native capabilities, see Anthropic's announcement on Claude 3.5 Sonnet and computer use. This shows how Claude can now control your computer and integrate with external systems, MCP is part of that broader ecosystem.

Comparing MCP to Alternative Integration Methods

You might be wondering: why MCP instead of other ways to connect SearchFit to Claude? Here's how MCP compares:

MCP vs. Manual Data Entry Manual copy-paste is slow and error-prone. MCP is automated, real-time, and eliminates human error.

MCP vs. API Calls in Custom Code If you're a developer, you could write custom code to call the SearchFit API and feed results to Claude. MCP is simpler, it handles the integration for you without requiring coding knowledge.

MCP vs. Web Scraping As outlined in guides on ethical web scraping with Claude, you could scrape SearchFit's dashboard. But that's fragile and violates terms of service. MCP uses official APIs, so it's reliable and authorized.

MCP vs. Plugins or Extensions Some tools offer browser plugins or extensions. MCP is native to Claude Desktop, so there's no separate extension to install or maintain.

In short, MCP is the cleanest, most reliable way to integrate external data into Claude.

Next Steps: Expanding Your AEO Workflow

Once you're comfortable using SearchFit in Claude, consider these next steps:

1. Audit Your Current AI Visibility Run a full audit using Claude and SearchFit data. Identify which platforms (ChatGPT, Perplexity, Gemini, Claude, Grok, Google) are driving the most citations and which are underperforming. Use this to prioritize your AEO efforts.

2. Analyze Your Competitor's AI Strategy Use Claude to compare your visibility to key competitors. Ask Claude to identify content gaps, topics where competitors are cited but you're not. This becomes your content roadmap.

3. Create an AEO Content Calendar Based on your SearchFit data and competitive analysis, build a content calendar optimized for AI search. Claude can help you prioritize topics, outline posts, and even draft content that's engineered to rank in AI engines.

4. Set Up Weekly Briefings Create a recurring prompt that Claude runs every Monday morning, showing you what changed in your AI visibility over the past week. This keeps your team aligned on trends and opportunities.

5. Integrate with Your Broader SEO Stack If you're using other SEO tools (like Semrush, SE Ranking, or HubSpot), consider how SearchFit data complements traditional SEO metrics. Claude can help you synthesize insights from multiple sources.

6. Test AEO Tactics Use Claude to brainstorm and test new AEO tactics. For example, ask Claude to suggest citation-worthy content formats, optimal content length for AI engines, or keyword strategies that align with how AI engines process queries.

Remember: AEO is still evolving. The platforms (ChatGPT, Perplexity, Gemini, Claude) are changing their citation and ranking logic constantly. By keeping SearchFit data close, literally in Claude, you'll stay agile and responsive to those changes.

Summary: You're Ready

You've now connected SearchFit to Claude Desktop. Here's what you've accomplished:

✓ Located and edited your Claude configuration file ✓ Installed the SearchFit MCP server package ✓ Configured your SearchFit API key ✓ Restarted Claude and tested the connection ✓ Learned how to query your AI visibility data directly from Claude ✓ Understood the advantages of MCP over other integration methods

Your AEO workflow is now streamlined. No more tab-switching. No more manual data pulls. You can ask Claude questions about your AI visibility and get answers backed by live SearchFit data.

The five-minute setup unlocks hours of time savings and enables a new level of strategic thinking. You're no longer just monitoring AI visibility, you're actively optimizing for it, with data-driven insights at your fingertips.

If you run into issues, refer back to the troubleshooting section. And if you want to explore more advanced use cases, like combining SearchFit data with Claude's web search capabilities or automating competitive intelligence, Claude is ready to help.

Start with a simple query: "What's my current AI visibility?" Then build from there. Your AEO strategy just got a lot more efficient.

Get SearchFIT Weekly

What we are learning about getting brands cited in AI search, in your inbox every week. Unsubscribe any time.