Updated June 2026 · MyoAmigo
An AI-ready strength training log
An AI-ready strength training log is a place to record your lifts that an AI assistant can actually read — your workouts, personal records, working-set volume, and stalled lifts — instead of a closed app whose data stays trapped behind charts. MyoAmigo is one. It's a local-first workout tracker for iPhone and Apple Watch, and its backend speaks the Model Context Protocol (MCP), the open standard Claude, ChatGPT, and Gemini use to reach real tools. Connect your assistant once, and it reads your real training history and writes routines and plans back into the app.
"AI workout tracker" sells the chatbot. This sells the data.
Most apps that advertise AI bolt a chatbot onto the logger and bill you for it. The model only sees what that vendor lets it see, and the advice never leaves their interface. An AI-ready strength training log inverts that: the log is the asset, and you point whatever assistant you already use at it. The intelligence is yours to choose; the data is the thing worth owning.
That distinction matters because the hard parts of training aren't logging the set — they're knowing whether a lift is genuinely stalling or just fluctuating, what to start a new exercise at, and whether today moved the needle. Those questions need your real history as context. A log built to hand that history to an assistant answers them; a chatbot guessing from a blank slate does not.
What "AI-ready" means in practice
MyoAmigo exposes 34 MCP tools, every one scoped to your authenticated account. They split into reads, analysis, and writes:
- Reads your real history: full workouts with set-level detail, per-exercise history, personal records, cardio, and body metrics — plus a dense context snapshot built for agents.
- Runs the app's analysis engines: estimated-1RM trends, a PR timeline, stall detection with a concrete prescription, the MyoMap per-muscle weekly heatmap, muscle volume against evidence-based bands, strength standards by bodyweight, honest session verdicts, and a coach-style week-in-review.
- Writes programming that lands: creates and revises routines and weekly plans, and logs sets and cardio with idempotency keys so a retry never double-logs. A drafted plan shows up in the app, scheduled and ready to run.
The machine-readable catalog of every tool lives at api.myoamigo.com/tools.json, and the remote MCP endpoint is api.myoamigo.com/mcp. The full reference is on the MCP server page.
The log itself
Before any AI touches it, MyoAmigo is a complete strength log. It works fully offline with no account required — every set lands in your phone's on-device database immediately, and your Apple Watch syncs directly with your iPhone with no cloud involved. Optional encrypted cloud sync is a separate switch you can turn on or revoke anytime. There's an 881-exercise built-in library plus custom exercises, and lossless Strong CSV import that keeps kilograms in kilograms. Weight is stored in the unit you entered it in, so a 100 kg set always reads "100 kg."
Things you can ask once it's connected
With your history available to the assistant, the prompts get specific:
- "What should I bench today based on my last 12 sessions?"
- "Why has my squat stalled, and what should I change?"
- "Build me a 4-day upper/lower block for the next eight weeks."
- "Which muscle groups am I under-training this week?"
- "What's a sensible starting weight for Romanian deadlifts — I've never logged them."
- "Summarize how this training block actually went."
The assistant answers from your real numbers, then can write the result back: a new routine, an updated weekly plan, or a logged set you dictated.
How an AI-ready log differs from a normal one
| Typical logger | MyoAmigo | |
|---|---|---|
| Where analysis lives | In-app charts only | In-app plus any MCP assistant |
| AI access to your data | None, or one vendor's chatbot | Your own assistant via MCP |
| Writes plans back | Manual | Assistant creates routines and plans in-app |
| Stall handling | You read the chart | Detection with a prescription |
| Data ownership | On the vendor's servers | On your phone; cloud sync optional |
Privacy
Access is account-gated and revocable anytime. Your training data is never used to train AI models and never sold. Nothing is shared by default, and nothing is shown to another user's agent.
Related reading
If you want the setup specifics for a particular assistant: use Claude as your workout tracker, give ChatGPT your real workout history, or read the developer-facing MCP server for fitness data page. Background: what is an MCP server and the AI personal trainer. The product overview is on the AI page.
Frequently asked questions
What is an AI-ready strength training log?
It's a lifting log built so an AI assistant can read your real training data — workouts, PRs, volume, and stalled lifts — and act on it, rather than keeping that data locked inside one app. MyoAmigo does this over MCP, the open protocol Claude, ChatGPT, and Gemini use to reach real tools.
Do I need to use AI to use MyoAmigo?
No. MyoAmigo is a full strength log on its own — it works offline with no account, with built-in analysis like stall detection and session verdicts. The MCP integration is optional and connects on top.
Which assistants can read my training log?
Any client that speaks MCP, including Claude, ChatGPT (via its connectors feature), and Gemini. You connect once at the remote endpoint api.myoamigo.com/mcp; access is gated by your account sign-in.
Is my training data used to train AI models?
No. Your data is never used to train models and never sold. Access is scoped to your account and revocable at any time.
