AI Agent Hub
Back to plugins
🤖

dsh-garmin-coach

Model Inference Updated 2026.09.05

Run the following command in DeepSeek Harness:

dsh plugin install csustyang/dsh-garmin-coach

Paste the following prompt into your AI chat to install this plugin:

Run dsh plugin install csustyang/dsh-garmin-coach in DSH to install the plugin, source code at https://github.com/csustyang/dsh-garmin-coach, then add the package to dsh.profile.bundles manually and restart DSH.

About this plugin

Garmin watches and sensors accumulate massive amounts of data every single day - pace, heart rate, HRV, cadence, TRIMP - yet most of it stays scattered across the Garmin Connect web portal, making it hard to turn raw numbers into actionable training decisions. dsh-garmin-coach bridges that gap by connecting a Garmin account to DeepSeek Harness, performing incremental data sync automatically, and letting an AI coach deliver insights grounded in your actual training history rather than generic template advice.

The core capability spans three layers: a sports dashboard with 30+ analytics (distance, pace, heart rate zones, HR-Pace, cadence, consistency, and more) that supports filtering and sorting by activity type; an AI training-advice engine that generates insights based on PB shifts, volume trends, and cadence deviations; and a structured training-plan generator with check-off tracking, version history, and a training diary for logging feelings and ratings. The plugin ships with 17 AI tools that can be invoked naturally in DSH conversations - simply ask for a monthly report, a personal best at 10K, or a half-marathon sub-2:00 plan, and the right tools are dispatched automatically.

All data stays on your local machine and is never uploaded to any server, making this a solid fit for runners, trail runners, and cyclists who want data-driven training decisions.

Use Cases

  • Review weekly pace trends and heart-rate zone distribution after a run
  • Ask the AI to generate next week's structured plan from the last 30 days of data
  • Query personal bests or monthly reports in plain natural language

Best For

  • Runners or cyclists who want data-driven training decisions
  • Garmin users who lack a structured training plan
  • Tech-savvy athletes who prefer querying sports analytics in natural language