Your workout log and strength coach, inside any AI assistant
Povver is a strength-training app for iOS. It reads every set you log, counts hard sets per muscle against the volume that muscle needs, and builds your next session from it. This MCP server gives Claude, ChatGPT, Gemini, or any MCP client read and write access to the same data, so you can work with your training log in a conversation instead of a screen: progressive overload, volume per muscle, plateaus, routines, periodization, injuries.
The server is hosted and remote. There is nothing to install. Every request is scoped to the account you sign in with, and it only touches your own data.
What you can ask
- "How is my bench press trending over the last 8 weeks? Am I still progressing or have I plateaued?"
- "Break down my weekly training volume by muscle group. Where am I under-training?"
- "What should I train today?"
- "Act as my strength coach: review my recent training, find my biggest weakness, and update my program to fix it."
- "Why did my squat weight drop last week? Explain the rule."
- "I tweaked my left shoulder. Record it so it stays out of my program."
- "Log this session for me: bench 4x6 at 80 kg, rows 3x10 at 60 kg."
- "Bring in my history from my old app."
- "Preview a periodization plan with a deload before you apply it."
What the assistant can do
Read. Povver does the math first. Weekly sets per muscle against the volume that muscle needs, fatigue, each lift's trend, bests and stalls, whether you're getting stronger: the assistant reads those finished numbers without adding up logs, so answers come quicker and use less of its context. The raw data is there for anything a summary misses: every workout down to each set, and your sets filtered by exercise or muscle. It also reads your routine and what's next, your injuries, your periodization plan and a session in progress.
Write. Build and edit routines and templates, and pick the active one. Log a workout you just did, or import history from another app. Log sets, swap an exercise or finish a session as it happens. Write a periodization plan with a deload, preview it, then apply it. Tell it about an injury so it stays out of your program. Accept or reject Povver's recommendations.
Delete. Workouts, routines and templates, as a separate action from editing.
Training rules. Ask why Povver counts sets or raises weight the way it does and the answer comes from the research behind the rule.
Resources you can attach as context: training status, your active routine, recent workouts, the latest analysis, and individual workouts and templates by ID.
Prompts: Analyze my training week, What should I train today?, Review exercise progress, Review my training program.
Connect
- Claude (Desktop and web): Settings, Connectors, Add custom connector, URL
https://mcp.povver.ai/mcp. Claude opens a browser window to sign in with your Povver account.
- ChatGPT (on a computer): add a custom MCP server with the same URL and choose OAuth. Setup: Connect ChatGPT to your workout log.
- Gemini (on a computer): add a custom connector with the same URL. Setup: Connect Gemini to your workout log.
- Claude Code:
claude mcp add --transport http povver https://mcp.povver.ai/mcp
- Cursor and other clients: add a remote server with the URL
https://mcp.povver.ai/mcp.
- Automation (scripts, n8n): in the Povver app go to More, Integrations, Developer access, create a key and send it as a Bearer token.
Setup and a worked conversation, from a stalling lift to a swapped routine: Your training log in Claude.
Security and privacy
OAuth 2.1 with PKCE and dynamic client registration. API keys are hashed at rest and revocable from the app. The server handles personal health data (training logs, body weight, injuries) and carries no sponsored content. The server source is closed. The public repository holds the documentation and the MCP Registry manifest.
Requirements
A Povver account with a subscription. Povver is an iOS app with a 14-day free trial that includes everything, then a monthly or yearly plan. There is no free tier. Current prices are in the app and on povver.ai.
Links
Comments