TrainingPeaks MCP server for Claude Desktop, Code and Cowork. No API approval needed - works with any account. Query workouts, CTL/ATL/TSB fitness data, power PRs via natural language.
Evidence-based AI coach for endurance training. Protocol-driven. Deterministic guidance for any LLM, with Intervals.icu integration.
The open-source European alternative to TrainingPeaks. Self-hostable endurance training platform. Your data stays yours.
🏃 An R package for advanced sports performance analysis and training load monitoring using Strava data.
Deterministic endurance coaching engine using Intervals.icu data, governed training logic, performance intelligence, adaptation progression and AI interfaces.
Own your training data! Open-source, self-hosted training log & analytics. Calendar, fitness/fatigue/form, dashboards, zones, peak curves & intervals.icu sync.
Local MCP server exposing your Intervals.icu training data to AI clients (Claude, etc.) — with server-side PMC / cardiac-decoupling math and a Stryd LBSS power extension.
🏋️♂️ Connect TrainingPeaks to AI assistants with ease. Query workouts, analyze data, and track fitness trends without API approval hassles.
Daily training-readiness score from Intervals.icu wellness data, pushed to your calendar.
Self-hosted MCP server on your own Garmin data: mirrors Garmin Connect into PostgreSQL, computes training load and muscle freshness, and serves it to Claude, ChatGPT and other clients over stdio or HTTP.
Free training hub for macOS: plan cycling, swimming, running and lifting; track estimated training loads, recovery and progress; preview programs with optional AI coaching via MCP. Runs locally, no Hub subscription. Includes Merach S29 smart-bike control.
Self-hosted cycling coach with a local LLM that never makes up a number — training load, overtraining detection & grounded AI advice.
The open-source engine behind journeytoironman.app — Strava + Apple Health to Postgres, sports-science KPIs (CTL/ATL/TSB, EF, ACWR), 16-week 70.3 plan, local + Databricks dashboards.
Evidence-based Claude Code skill: Garmin FR570 → 4-tier daily training decision + 8-KPI weekly fat-loss review. 206 citations, deterministic decision tree, anti-overreach red lines.
Este repositorio recoge el código para el proyecto Trail Analytics: Exploración visual y modelado predictivo de carreras por montaña.
Curated research on training periodization: linear, undulating, block, conjugate, and polarized models. Tapering science, training load monitoring, and individualization evidence.
Interpreting wearable fitness data through a physiology lens: HRV, VO2 max estimation, sleep tracking accuracy, training load metrics, and device validation studies.
Local, self-hosted training log with Garmin sync, fitness/training-load analytics, and an AI coach: fart (speed) + lek (play).
Vendor-neutral, local-first and auditable sports and recovery trends with deterministic metrics and an evidence-gated AI coach core.
Private, data-first training dashboard for your WHOOP data: sports-science load metrics, explainable daily call, and a full-history AI chat. Local-first, zero dependencies.
To associate your repository with the training-load topic, visit your repo's landing page and select "manage topics."