Runalyze MCP server gives AI the richest sports dataset yet.

Runalyze, the German training analysis platform used by serious endurance athletes, launched an official MCP server on 9 June. MCP, the Model Context Protocol, is the glue between a chat app and your personal data, letting an AI tool read your Runalyze account directly once you permit it. It is open only to paying members and is still in testing. You can connect on the Runalyze settings page.

The launch follows COROS in May and Strava in June.

COROS opened a single source: activities recorded on a COROS watch. Strava is different. It already receives uploads from Garmin, Wahoo, COROS, and dozens of other devices, so its MCP draws on whatever an athlete has synced, not just Strava-native recordings. Runalyze works the same way and goes further. Athletes use it to bring everything together, feeding in from Garmin, Polar, Suunto, COROS, Whoop, Oura, and more. Where Strava shares activity and route data to AI, Runalyze adds a full health layer on top!

The data on offer is the broadest of any sports MCP so far, training and health combined in one place. Activities, splits, and VO2max estimates sit alongside training metrics: CTL, ATL, TSB, acute:chronic ratio, training monotony and strain, marathon shape, HRV baseline and normal range. The health data adds sleep stages and quality, overnight heart rate, respiratory rate, resting heart rate trends, weight, body composition, blood glucose, blood pressure, body temperature, stress, mood, fatigue, daily notes, injuries, and illness logs.

An athlete who has used Runalyze for three years and connected every device they own is handing Claude an unusually detailed longitudinal training record.

The questions Claude can answer go well beyond workout summaries. How did HRV respond during the heaviest training weeks of a marathon build? What changed before a personal best? Is fatigue rising faster than fitness? Runalyze already answers these questions inside its own interface. The MCP makes the same answers available in plain conversation, with no dashboards to navigate.

That is where the risk to Runalyze lies, but it is smaller than it first appears. Runalyze’s outputs are calculated figures, rather than raw data. The MCP hands over what Runalyze has already calculated: the VO2max estimate, the TSB figure, and the HRV baseline. Claude gets the answer, not the method behind it. Runalyze maintains control over how its numbers are produced, the same position COROS took when it shared summary training load through its MCP but withheld per-second GPS and heart rate data.

The real pressure falls elsewhere. Coaching platforms in the TrainingPeaks category exist mainly to turn training data into recommendations. They now face an AI assistant that can perform a version of the same job on a broader dataset and explain its reasoning in ordinary language. That is a harder position to defend than Runalyze’s.

Hardware makers are not threatened at all. No MCP connection, however complete, can produce a GPS track, a heart rate reading, or a power figure. That core data must always come from a physical device. AI competes with the analysis, not the recording. The device stays essential, but its role narrows. If the insight moves to a chat tool that reads from any brand equally, the watches start to look interchangeable, and the lock-in that came from each company’s own app weakens.

The athletes most likely to connect to this quickly are those already using Runalyze. They run multiple devices, perhaps from multiple device platforms.

Across COROS, Strava, and now Runalyze, sports platforms are opening their data to AI faster than most would have predicted a year ago. Each connection widens what AI can reason about.

Takeaway

Once AI can ‘correctly’ reason, apps and platforms will disappear. Once hardware vendors open up to AI they become swappable.

That is the essence of the threat facing the entire industry.

Last Updated on 22 June 2026 by the5krunner

the5krunner

tfk is the founder and author of the5krunner, an independent endurance sports technology publication. With 20 years of hands-on testing of GPS watches and wearables, and competing in triathlons at an international age-group level, tfk provides in-depth expert analysis of fitness technology for serious athletes and endurance sport competitors. ID