How To Connect LiftTrack to Your AI Assistant
Use AI to analyze your workouts, check training maxes, and plan your training by connecting LiftTrack to ChatGPT, Claude, Cursor, and more
You can now connect LiftTrack directly to AI assistants like ChatGPT and Claude. Instead of manually digging through your workout history, just ask your AI a question in plain English: “How has my bench press progressed this month?” or “What’s on my training schedule this week?”
This guide walks you through connecting LiftTrack to the most popular AI clients so your assistant has full access to your workout data.
What is MCP?
MCP (Model Context Protocol) is an open standard that lets AI assistants connect to external apps and services. Think of it like a universal adapter: instead of every AI tool building its own custom integration, MCP provides one standard way for AI assistants to securely read data from the apps you already use.
LiftTrack’s MCP server gives your AI assistant direct access to your workout data, so you can have natural conversations about your training without copy pasting numbers or switching between apps.
Coming Soon: Official Apps
An official LiftTrack integration for Claude and a ChatGPT app are currently being submitted for review. Once approved, connecting will be even simpler with a single click from their respective app directories.
In the meantime, you can connect manually using the steps below. It only takes a minute.
What You Can Ask Your AI
Once connected, you can ask things like:
- “What did I lift last week?” — Get a summary of all your completed sessions, including exercises, sets, weights, and duration.
- “Show me my bench press progress” — See your performance history for any exercise across sessions, including estimated 1RM trends and volume.
- “What’s my training max for squat?” — Check your current training max values used for percentage based programs like 5/3/1.
- “What workouts do I have scheduled this week?” — View your upcoming scheduled workouts.
- “Search for glute exercises” — Browse the full exercise catalog by name or muscle group.
- “Break down my Push Pull Legs routine” — Get the full detail of any workout template including sets, weights, rest times, superset grouping, and RPE targets.
Your data is read only. Your AI assistant can view your workouts and history but cannot modify anything in your account.
Setup Guides
ChatGPT
Requirements: ChatGPT Plus, Pro, or Team subscription
- Open chatgpt.com and click your profile icon in the top left.
- Select Settings then Apps and Connectors.
- Scroll to the bottom and select Advanced Settings.
- Toggle Developer Mode to on.
- Go back to Apps and Connectors and click Create.
- Fill in the details:
- Name: LiftTrack
- Description: Access your LiftTrack workout data, training history, training maxes, and exercise catalog.
- Connector URL:
https://mcp.lifttrackapp.com/mcp
- Click Create.
- When prompted, sign in with your LiftTrack account.
You can now ask ChatGPT about your workouts in any conversation.
Claude Desktop
- Open the Claude desktop app and go to Settings then Connectors.
- Click Add custom connector at the bottom of the page.
- Enter the URL:
https://mcp.lifttrackapp.com - Click Add.
- Click Configure next to LiftTrack and sign in with your LiftTrack account.
To use LiftTrack in a conversation, click the + button in the lower left of the chat, then select Connectors and toggle LiftTrack on.
Claude Code (CLI)
Run the following command in your terminal:
claude mcp add lifttrack --transport http https://mcp.lifttrackapp.com
Claude Code will prompt you to sign in on first use.
Cursor
- Open Cursor and go to Settings then Tools & Integrations.
- Click Add MCP Server.
- Enter the URL:
https://mcp.lifttrackapp.com
Alternatively, add this to your ~/.cursor/mcp.json file:
{
"mcpServers": {
"lifttrack": {
"url": "https://mcp.lifttrackapp.com"
}
}
}
Gemini CLI
Run the following command in your terminal:
gemini mcp add --transport sse lifttrack https://mcp.lifttrackapp.com
Alternatively, add this to your ~/.gemini/settings.json file:
{
"mcpServers": {
"lifttrack": {
"url": "https://mcp.lifttrackapp.com"
}
}
}
Signing In
On first use, your AI client will open a browser window where you sign in with your LiftTrack account (the same email and password you use in the app). After signing in, your session stays active so you won’t need to do this again for a while.
If your session expires, the AI client will prompt you to sign in again automatically.
Example Conversations
Here are a few ways you might use LiftTrack with your AI assistant:
Review your training week
You: How did my training go this past week?
AI: You completed 4 sessions this week. Here’s a quick breakdown…
- Monday: Upper Body (45 min) — Bench Press 4x8 at 185 lbs, Barbell Row 4x8 at 155 lbs…
- Wednesday: Lower Body (52 min) — Back Squat 5x5 at 275 lbs, Romanian Deadlift 3x10 at 185 lbs…
Total volume was up 8% compared to last week. Nice work.
Check if you’re ready to go heavier
You: Should I increase my bench press weight? Show me my recent progress.
AI: Looking at your last 10 bench press sessions, you’ve hit all your prescribed reps at 185 lbs for the past 3 sessions. Your estimated 1RM has been trending up steadily from 225 to 232 lbs. Looks like you’re ready to move up. Based on your rounding settings, I’d suggest trying 190 lbs next session.
Plan around your schedule
You: What do I have coming up this week and is there anything I should adjust?
AI: You have 3 workouts scheduled:
- Tuesday: Push Day
- Thursday: Pull Day
- Saturday: Leg Day
Based on your recent history, you’ve been consistent with this split. Your pull day volume has been higher than your push day — you might consider adding an extra set of overhead press on Tuesday to balance things out.
Tips
- Be specific with exercise names. “Barbell Bench Press” will get better results than just “bench.” If you’re unsure of the exact name, ask the AI to search the exercise catalog.
- Use date ranges. When asking about history, specify a time frame like “last 2 weeks” or “in January” for more focused answers.
- Try it for programming ideas. Ask the AI to look at your current routine and suggest modifications based on your actual performance data.
Troubleshooting
- “Not authenticated” error: Your session may have expired. The AI client should prompt you to sign in again. If not, try removing and re adding the LiftTrack MCP server.
- Data looks outdated: The MCP server reads directly from your account in real time. Make sure your recent workouts have synced from your Garmin watch or were logged in the app.
- AI doesn’t use LiftTrack tools: Some AI clients need you to explicitly mention LiftTrack or your workout data for the assistant to use the connection. Try being specific: “Using LiftTrack, show me my squat history.”