Integrations
Gravl MCP
The Gravl MCP server lets AI assistants act on your account through the Model Context Protocol — log a workout by describing it, tweak your split mid-conversation, ask what you benched last month. Scopes are enforced on every call, and the connection is revocable any time.
Connect
Add Gravl as a custom connector in your AI app with this server URL:
https://mcp.gravl.ai/mcpClaude
- Open Settings → Connectors and choose “Add custom connector”.
- Name it Gravl and paste the server URL.
- Choose Connect and sign in with your Gravl account.
ChatGPT
- Open Settings → Apps & Connectors → Advanced settings and turn on Developer mode.
- Choose Create, name it Gravl, paste the server URL and pick OAuth.
- Sign in with your Gravl account when ChatGPT asks.
Grok Bot
- Ask the bot: “Add a custom MCP server called Gravl at
https://mcp.gravl.ai/mcp”. - Check the URL and choose Add it.
- Choose Authorize on the connect card and sign in with your Gravl account.
Muse
- Ask Muse to add a custom connector for Gravl at
https://mcp.gravl.ai/mcp. - Open the sign-in link Muse sends and sign in with your Gravl account.
Other apps that support remote MCP servers with OAuth connect the same way.
You sign in on Gravl’s own page with your Gravl account — the same email and password, Google or Apple login you use in the app. Connecting never creates an account, and the assistant only ever sees your own data.
Try asking
- “Log today’s workout: bench press 4×8 at 80 kg, then 3 sets of pull-ups to failure.”
- “How did my squat change over the last three months?”
- “Swap barbell squats for leg press in my Leg Day template.”
- “Build me a 4-day upper/lower split for my home gym.”
What it can do
Log & manage workouts
Log, read and delete workouts. Logged workouts get volume, calories, PRs and trophies exactly like the app.
Templates & splits
Create and edit workout templates and custom splits, or switch to one of Gravl’s own splits.
Progress & records
Training summaries, exercise history, personal records, strength scores and body measurements.
Training settings
Change goals, muscle focus, warmup, rest times and workout preferences, one field at a time.
Gyms & equipment
Read your gym profiles and update the equipment Gravl plans your workouts around.
Exercise catalog
Search exercises, find alternatives, and add a custom exercise when a movement is missing.
Workouts logged through MCP are first-class: they get volume and calorie calculations, personal records, trophies and feed entries exactly as if logged in the app, and the app picks them up on its next sync.
Security model
| Layer | What it does |
|---|---|
| Gravl sign-in | You sign in on Gravl’s own page, never inside the AI app. The AI app never sees your password. |
| Per-connection scopes | Each connection carries an explicit scope set; every tool call is checked against it server-side. |
| Account linking | Each connection is tied to the Gravl account you sign in with. It never creates a new account. |
| Revocation | Disconnect from your MCP client, or revoke the connection from your Gravl account — either kills access immediately. |
MCP or the REST API?
| You’re building… | Use |
|---|---|
| An AI assistant experience (chat, agents) on your own account | MCP |
| A dashboard, script or integration reading your data | A personal token + REST API |
| An app other Gravl users sign in to | OAuth + REST API |