Strava MCP Server
# Strava MCP Server
A Model Context Protocol (MCP) server that connects Claude to your Strava account. Ask Claude in natural language to analyze your training, predict race times, compute training loads, or generate a full periodized training plan.
## Features
- **13 tools** across 5 categories: auth, activities, analysis, prediction, planning
- OAuth2 flow with automatic local callback server — no manual code copying
- VDOT-based training paces (Jack Daniels' Running Formula)
- CTL/ATL/TSB training load metrics (Chronic/Acute Training Load, Training Stress Balance)
- Race time predictions via Riegel formula
- Full periodized training plans (Base → Build → Peak → Taper) with race-specific logic
## Requirements
- Node.js ≥ 18
- A Strava account
- Claude Desktop (or any MCP-compatible client)
---
## Setup
### 1. Create a Strava API application
Go to [strava.com/settings/api](https://www.strava.com/settings/api) and create an application.
- **Authorization Callback Domain**: `localhost`
Note your **Client ID** and **Client Secret**.
### 2. Configure environment variables
```bash
cp .env.example .env
```
Edit `.env`:
```env
STRAVA_CLIENT_ID=your_client_id
STRAVA_CLIENT_SECRET=your_client_secret
STRAVA_REDIRECT_URI=http://localhost:8080/callback
TOKENS_FILE_PATH=./tokens.json
```
### 3. Build
```bash
npm install
npm run build
```
### 4. Configure Claude Desktop
Edit `%APPDATA%\Claude\claude_desktop_config.json` (Windows) or `~/Library/Application Support/Claude/claude_desktop_config.json` (macOS):
```json
{
"mcpServers": {
"strava": {
"command": "node",
"args": ["C:/path/to/McpStrava/dist/index.js"],
"env": {
"STRAVA_CLIENT_ID": "your_client_id",
"STRAVA_CLIENT_SECRET": "your_client_secret",
"TOKENS_FILE_PATH": "C:/path/to/McpStrava/tokens.json"
}
}
}
}
```
> You can use either `.env` or the `env` block in the Claude Desktop config — both work.
### 5. Authenticate
Restart Claude Desktop, then in a conversation:
1. Call `strava_get_auth_url` — Claude will return a URL
2. Open the URL in your browser and authorize the app on Strava
3. The page will show **"✓ Authentification réussie !"** and tokens are saved automatically
---
## Available Tools
### Authentication
| Tool | Description |
|------|-------------|
| `strava_get_auth_url` | Generates the OAuth2 URL and starts the local callback server |
| `strava_exchange_token` | Manual fallback: exchange an auth code for tokens |
| `strava_auth_status` | Check if tokens are valid and when they expire |
### Activities
| Tool | Description |
|------|-------------|
| `strava_get_activities` | List recent activities (Run, Ride, Walk, All) with distance, pace, HR |
| `strava_get_activity_detail` | Full detail for one activity: splits per km, laps, calories |
| `strava_athlete_stats` | Global Strava stats: this week, this year, all-time |
### Analysis
| Tool | Description |
|------|-------------|
| `strava_analyze_training` | Weekly volume breakdown + consistency score |
| `strava_training_load` | CTL (fitness) / ATL (fatigue) / TSB (freshness) via TRIMP |
| `strava_pace_zones` | Distribution across 6 pace zones, 80/20 rule check |
### Prediction
| Tool | Description |
|------|-------------|
| `strava_predict_race_time` | Predict finish times via Riegel formula from a reference effort |
| `strava_vdot` | Compute VDOT score + 5 training pace zones from any race performance |
### Training Planning
| Tool | Description |
|------|-------------|
| `strava_generate_training_plan` | Full periodized plan from today to race day (Base/Build/Peak/Taper) |
| `strava_weekly_workout` | Generate just next week's sessions for a given phase |
---
## Training Plan Details
### Phases
| Phase | Focus | Intensity |
|-------|-------|-----------|
| Base | Aerobic foundation | Easy runs, long run, strides |
| Build | Lactate threshold | Tempo, easy medium, long run |
| Peak | VO2max + race-specific | Intervals, tempo, long run |
| Taper | Freshness | Reduced volume, maintenance quality |
### How volume is calculated
1. **Starting volume** — blend of your 4-week average km and CTL-derived weekly km estimate (robust to injury breaks)
2. **Peak volume** — 1.4× current, with race-specific minimums (5K: 40km, 10K: 50km, Half: 60km, Marathon: 80km)
3. **Weekly progression** — capped at +10% per week to prevent injury
4. **Recovery weeks** — automatic every 4th week within each phase (volume × 0.8)
### Race-specific logic
- **Marathon (Build & Peak)**: Long runs include a marathon-pace section (~45% of the run at race pace)
- **Taper depth**: 5K tapers to 80% of peak volume, Marathon to 40% — shorter races need less recovery
- **Intervals**: 6 × 1000m for 5K/10K, 5 × 1000m for Half/Marathon
### VDOT & pace zones
Based on Jack Daniels' Running Formula. Zones computed as a fraction of VDOT:
| Zone | % of VDOT | Use |
|------|-----------|-----|
| Easy | 65% | Daily runs, long run |
| Marathon | 80% | Marathon-pace sections |
| Threshold | 86% | Tempo runs |
| Interval | 98% | VO2max intervals |
| Repetition | 105% | Speed work / strides |
---
## Generating plans for friends (manual mode)
`strava_generate_training_plan` supports a **manual mode** that bypasses Strava entirely. Pass `current_weekly_km` and `goal_time` together and no Strava account is needed — useful for generating plans for friends from your own Claude Desktop.
### Required parameters
| Parameter | Description | Example |
|-----------|-------------|---------|
| `target_race` | Race distance | `"Marathon"` |
| `race_date` | Race date (YYYY-MM-DD) | `"2026-10-18"` |
| `goal_time` | Target finish time | `"3:45:00"` |
| `current_weekly_km` | Current weekly mileage | `55` |
### Example prompts
```
Génère un plan marathon pour mon ami, il court 55km par semaine et vise 3h45, la course est le 18 octobre 2026
```
```
Mon amie veut courir un semi-marathon en 1h50 le 2026-09-14, elle fait environ 40km par semaine
```
When both `current_weekly_km` and `goal_time` are provided, the tool skips all Strava API calls. The `resume` in the response will include `source_volume: "Fourni manuellement"` to confirm which mode was used.
### Mode comparison
| | Strava mode | Manual mode |
|---|---|---|
| Strava account needed | Yes | No |
| Volume calibration | 4-week avg + CTL | Value you provide |
| VDOT estimation | From recent activities or `goal_time` | From `goal_time` (required) |
| Use case | Your own training | Friends / athletes without Strava |
---
## Development
```bash
# Watch mode (no build step needed)
npm run dev
# Build TypeScript
npm run build
# Run built server
npm start
# Clean build artifacts
npm run clean
```
### Project structure
```
src/
├── index.ts # MCP server entry point
├── config.ts # Env vars, Strava constants, race distances
├── types.ts # Shared TypeScript interfaces
├── auth/
│ ├── oauth.ts # OAuth2 URL builder, token exchange
│ ├── tokenStore.ts # Load/save tokens.json, expiry check
│ ├── callbackServer.ts # Local HTTP server for OAuth redirect
│ └── authTools.ts # MCP auth tools
├── strava/
│ ├── client.ts # Axios instance with auto token refresh
│ ├── activities.ts # Strava activities API
│ ├── athlete.ts # Strava athlete/stats API
│ └── activityTools.ts # MCP activity tools
├── analytics/
│ ├── metrics.ts # Weekly stats, pace zones, consistency score
│ ├── trainingLoad.ts # TRIMP, CTL/ATL/TSB computation
│ └── analysisTools.ts # MCP analysis tools
├── prediction/
│ ├── riegel.ts # Riegel race time prediction formula
│ ├── vdot.ts # VDOT computation, training paces, race equivalents
│ └── predictionTools.ts # MCP prediction tools
└── planning/
├── workouts.ts # Workout templates and distance bounds
├── plan.ts # Plan generation, phase allocation, VDOT estimation
└── planTools.ts # MCP planning tools
```
### Tokens
`tokens.json` stores your Strava access and refresh tokens. It is in `.gitignore` — never commit it. Tokens are refreshed automatically when they expire (Strava access tokens last 6 hours).
---
## Example prompts
**Strava mode (your own account)**
```
Analyse mes 8 dernières semaines d'entraînement
```
```
Quelle serait mon heure sur un marathon si je cours un 10K en 45min ?
```
```
Génère-moi un plan d'entraînement pour un semi-marathon le 2026-09-20
```
```
Calcule ma charge d'entraînement actuelle et dis-moi si je suis en forme pour une course ce week-end
```
```
Montre-moi la répartition de mes allures sur les 4 dernières semaines
```
**Manual mode (friends / no Strava)**
```
Génère un plan marathon pour mon ami, il court 55km par semaine et vise 3h45, la course est le 18 octobre 2026
```
```
Mon amie veut courir un semi-marathon en 1h50 le 14 septembre 2026, elle fait 40km par semaine
```
---
## License
MIT
TDQS
Scored across 13 tools
Most tools target distinct aspects of training analysis (load, pace zones, VDOT, weekly plan), but some overlap exists between race time prediction tools (predict_race_time and vdot both estimate race times, though from different formulas). Descriptions help differentiate, but potential confusion remains.
All tools start with 'strava_' and use snake_case, but there is inconsistency in verb-noun order: some start with a verb (strava_analyze_training, strava_get_activities) and others with a noun (strava_athlete_stats, strava_pace_zones). This mixed pattern reduces predictability.
13 tools is a well-scoped set for a training analysis server. It covers authentication, activity retrieval, multiple analysis dimensions, and plan generation without being overwhelming.
The tool set provides comprehensive coverage of training analysis (load, pace, VDOT, race prediction, plan generation) and basic activity retrieval. Minor gaps exist, such as no tool to fetch or modify generated training plans, but core workflows are supported.