Overview
The Model Context Protocol (MCP) bridges your personal Concept2 logbook history and MyNextRow's physiological coaching models directly into AI applications like Claude Desktop and Cursor.
Once configured, you no longer need to type out your split times, past distance, or fatigue scores into prompts. Your AI assistant will automatically query MyNextRow to inspect your training metrics, verify your Concept2 connection, and generate structured workouts.
Step 1: Generate Your MyNextRow API Key
Before configuring your desktop assistant, you need an authenticated API key:
- Log into your account at mynextrow.com.
- Open your account Settings at mynextrow.com/settings.
- Scroll down to the Model Context Protocol (MCP) & API Keys card.
- Click Create API Key.
- Give your key a descriptive name (for example,
Claude DesktoporCursor). - Click Generate Key.
- Important: Copy your newly generated key immediately (
mnr_live_...). For your security, raw API keys are never shown again after this dialog is dismissed.
Step 2: Configure Claude Desktop
Claude Desktop uses a JSON configuration file to launch local MCP server processes.
1. Locate Your Configuration File
- macOS:
~/Library/Application Support/Claude/claude_desktop_config.json - Windows:
%APPDATA%\Claude\claude_desktop_config.json(usuallyC:\Users\<YourUsername>\AppData\Roaming\Claude\claude_desktop_config.json)
Tip: If the file or folder does not exist yet, you can create it.
2. Add the MyNextRow MCP Server
Open claude_desktop_config.json in your favorite text editor and add the mynextrow entry under mcpServers:
{
"mcpServers": {
"mynextrow": {
"command": "npx",
"args": [
"-y",
"@mynextrow/mcp-server"
],
"env": {
"MYNEXTROW_API_KEY": "YOUR_COPIED_API_KEY_HERE"
}
}
}
}
Replace YOUR_COPIED_API_KEY_HERE with the actual key you generated in Step 1.
3. Restart Claude Desktop
Save the file and quit Claude Desktop completely (Cmd + Q on macOS or right-click the tray icon and exit on Windows). When you relaunch Claude Desktop, you will see a small hammer or tool icon in the bottom right corner of the input area indicating that the MyNextRow tools are active.
Step 3: Configure Cursor IDE
If you use Cursor for code or research, you can enable MyNextRow's coaching tools in Cursor's Composer and Chat interface:
- Open Cursor and navigate to Settings (
Cmd + ,on macOS orCtrl + ,on Windows). - In the left sidebar, click Features, then scroll to MCP Servers.
- Click Add New MCP Server.
- Configure the fields:
- Name:
mynextrow - Type:
command(orstdio) - Command:
npx -y @mynextrow/mcp-server
- Name:
- In the environment variables section, add:
MYNEXTROW_API_KEY:YOUR_COPIED_API_KEY_HERE
- Click Save and verify the green status indicator turns on.
Step 4: Test Your Connection
Once connected, you can converse naturally with your AI assistant. Here are a few recommended queries to verify your connection:
Checking Your Training Load
"Claude, can you check my current rowing training load and injury risk using MyNextRow?" Claude will call
get_training_load(0 credits) and return your Chronic Training Load (fitness), Acute Training Load (fatigue), and ACWR.
Reviewing Recent Logbook Sessions
"What were my last three rowing sessions on Concept2, and what was my average pace?" Claude will call
get_recent_workouts(0 credits) to summarize your recent distance and split averages.
Generating an Adaptive Workout
"Please design a 40-minute aerobic threshold rowing session tailored to my current 2k pace." Claude will call
generate_workout(1 credit) to build a structured interval workout with warm-up, main work blocks, stroke rates, and target pace clamps, save it directly to your MyNextRow library, and provide an instant ErgData PM5 direct deep link so you can program your monitor in one click.
Built-in Prompts
Claude Desktop and Cursor also support built-in prompt templates:
plan_todays_row: Automatically checks your fatigue metrics and suggests today's optimal training piece.review_recent_progress: Evaluates your pacing consistency and technique streak over recent weeks.
Security Best Practices
- Keep your keys private: Never commit your
claude_desktop_config.jsonor any file containingMYNEXTROW_API_KEYto public GitHub repositories or share them in screenshots. - Use the in-app masking toggle: In MyNextRow settings, the configuration snippet includes a "Mask key for sharing" button that replaces your live secret with
<PASTE_YOUR_API_KEY_HERE>. - Revoking keys: If you ever suspect a key has been compromised, open MyNextRow Settings and click the trash can icon next to the key to revoke it immediately.
Key points
- Generate a secure 256-bit API key from the MyNextRow Settings page.
- Configure Claude Desktop by editing claude_desktop_config.json on macOS or Windows.
- Configure Cursor IDE through Settings > Features > MCP Servers.
- Test your setup with sample prompts like checking training load or generating a workout.
- Protect your credentials by never committing client configuration files to public repositories.
Editorial & Coaching Standards
Guides in the MyNextRow Learn library are created by Concept2 athletes and coaches, synthesizing peer-reviewed sports physiology, biomechanics literature, and authoritative rowing guidelines. Every workout protocol and technical cue is tested for safety, repeatability, and PM5 monitor compatibility.
Sources and further reading
- Claude Desktop — Model Context Protocol Quickstart— Official setup documentation for configuring MCP servers in Claude Desktop.
- Cursor — MCP Integration Documentation— Reference guide for adding and configuring stdio MCP servers in Cursor IDE.
- MyNextRow — Settings and API Keys— In-app account dashboard for generating, masking, and revoking MCP API keys.