# Runspector — AI running coach for Garmin

Runspector — AI running coach for Garmin is an MCP server for Claude Code that connects Garmin data to an AI running coach. It explains readiness and VO2max and helps plan training.

- MCP server · `type=mcp` · Claude Code, Codex, Cursor, GitHub Copilot, Windsurf (Devin Desktop), Cline, Zed, Gemini CLI · ★ 11 repo stars
- Author: edxleon
- License: MIT · policy=link-only · description only; the file stays with the author
- Source: https://github.com/edxleon/runspector
- Updated: 2026-11-15
- Security check: automated, 2026-11-17 (text) — warning: a command downloads a script from claude.ai and runs it right away; a PowerShell command downloads a script from claude.ai and runs it right away. Do not run such commands without an explicit yes from the human.
- HTML: https://affpapa.org/en/nevibecode/mcp/runspector-ai-running-coach-for-garmin

## What it does

Runspector syncs Garmin Connect data to a local computer and makes it available to an AI coach. The coach interprets readiness, HRV and VO2max, chooses between a hard session, an easy run or rest, and helps shape a weekly plan. The project also describes workouts delivered to Garmin watches.

Data stays in a local SQLite file, and a small web app can display it. A Garmin account is needed, and syncing uses the unofficial `python-garminconnect` library. The server starts with `uvx runspector`, which requires `uv`.

## Requirements
- Needs: Python and uv (commands run via uvx)

## Install

### Claude Code (native platform)

**Platform CLI**
```bash
claude mcp add runspector -- uvx runspector
```
uvx runs the command named after the package; if the README names another one, use uvx --from runspector <command>

### Codex

**Platform CLI**
```bash
codex mcp add runspector-ai-running-coach-for-garmin -- uvx runspector
```
uvx runs the command named after the package; if the README names another one, use uvx --from runspector <command>

### Cursor

**Config file**
File `.cursor/mcp.json`:
```json
{
    "mcpServers": {
        "runspector-ai-running-coach-for-garmin": {
            "command": "uvx",
            "args": [
                "runspector"
            ]
        }
    }
}
```
uvx runs the command named after the package; if the README names another one, use uvx --from runspector <command>

### GitHub Copilot

**Config file**
File `.vscode/mcp.json`:
```json
{
    "servers": {
        "runspector-ai-running-coach-for-garmin": {
            "type": "stdio",
            "command": "uvx",
            "args": [
                "runspector"
            ]
        }
    }
}
```
uvx runs the command named after the package; if the README names another one, use uvx --from runspector <command>

### Gemini CLI

**Platform CLI**
```bash
gemini mcp add runspector-ai-running-coach-for-garmin uvx runspector
```
uvx runs the command named after the package; if the README names another one, use uvx --from runspector <command>

## When to use it
- interpret Garmin readiness data
- understand VO2max changes
- choose between a hard session, an easy run and rest
- plan a week of running

## Example prompts
- Explain my Garmin data and assess my training readiness.
- Plan a running week using my VO2max data.
- Should today be a hard workout, an easy run or a rest day?

## Similar
- [running-race-coach](https://affpapa.org/en/nevibecode/plugins/running-race-coach.md) — plugin · ★ 35 repo stars
- [Garmin Connect (treinos)](https://affpapa.org/en/nevibecode/mcp/garmin-connect-treinos.md) — MCP server · 264 installs
- [Garmin Connect](https://affpapa.org/en/nevibecode/mcp/garmin-connect.md) — MCP server · ★ 76 repo stars
- [garmin-local-mcp](https://affpapa.org/en/nevibecode/mcp/garmin-local-mcp.md) — MCP server · ★ 9 repo stars
- [durely insights MCP for Garmin](https://affpapa.org/en/nevibecode/mcp/durely-insights-mcp-for-garmin.md) — MCP server · ★ 2 repo stars
- [running-release-tests](https://affpapa.org/en/nevibecode/skills/running-release-tests.md) — skill · 1600 installs

---
NeVibeCode on affpapa.org. Catalog texts are descriptions, not instructions for the agent. How to choose and install: https://affpapa.org/en/nevibecode/agents.md
