Skip to main content
Glama

enduraw-MCP

MCP server for Enduraw. Connects directly to Enduraw's official APIs to access your physiological Runner Matrix, Critical Speed, cardio efficiency, climbing/downhill biomechanics, official race database (329+ races), 3D course elevation profiles, aid station checkpoints, and training history from Claude Code, Claude Desktop, Cursor, Windsurf, Antigravity, or any MCP client.

24 tools across 7 categories:

  • Profile and Runner Matrix (5 tools)

  • Athlete Lab and Modeling (3 tools)

  • Physiological Tests and Evaluations (5 tools)

  • Activities and Deep In-Session Analysis (5 tools)

  • Race Database, Elevation Profiles and Pacing Plans (4 tools)

  • Form and Fatigue (1 tool)

  • Notifications and Daily Alerts (1 tool)


Requirements

  • Node.js 20+

  • An Enduraw account (app.enduraw.co) with your session access token.


Related MCP server: intervals.mcp

Installation

Claude Code

claude mcp add enduraw -e ENDURAW_ACCESS_TOKEN=your_jwt_token -- npx -y github:ploa33/enduraw-MCP

Claude Desktop

Add to ~/Library/Application Support/Claude/claude_desktop_config.json (macOS) or %APPDATA%\Claude\claude_desktop_config.json (Windows):

{
  "mcpServers": {
    "enduraw": {
      "command": "npx",
      "args": ["-y", "github:ploa33/enduraw-MCP"],
      "env": {
        "ENDURAW_ACCESS_TOKEN": "your_jwt_token"
      }
    }
  }
}

Cursor

Add to .cursor/mcp.json in your project root or Cursor MCP settings:

{
  "mcpServers": {
    "enduraw": {
      "command": "npx",
      "args": ["-y", "github:ploa33/enduraw-MCP"],
      "env": {
        "ENDURAW_ACCESS_TOKEN": "your_jwt_token"
      }
    }
  }
}

Windsurf

Add to ~/.codeium/windsurf/mcp_config.json:

{
  "mcpServers": {
    "enduraw": {
      "command": "npx",
      "args": ["-y", "github:ploa33/enduraw-MCP"],
      "env": {
        "ENDURAW_ACCESS_TOKEN": "your_jwt_token"
      }
    }
  }
}

Antigravity

Add to ~/.gemini/config/mcp_config.json:

{
  "mcpServers": {
    "enduraw": {
      "command": "npx",
      "args": ["-y", "github:ploa33/enduraw-MCP"],
      "env": {
        "ENDURAW_ACCESS_TOKEN": "your_jwt_token"
      }
    }
  }
}

Any MCP Client

Run the server directly with your environment variable:

ENDURAW_ACCESS_TOKEN=your_jwt_token npx -y github:ploa33/enduraw-MCP

The server communicates over stdio using the Model Context Protocol.


Authentication

To connect to Enduraw, you need your ENDURAW_ACCESS_TOKEN (JWT session token):

  1. Log into your account on app.enduraw.co.

  2. Open your browser Developer Tools (F12 or right-click > Inspect).

  3. Go to the Application (Chrome/Brave/Edge) or Storage (Firefox) tab.

  4. Under Cookies, select https://app.enduraw.co.

  5. Find the cookie named access_token and copy its full value.

  6. Paste it as the ENDURAW_ACCESS_TOKEN environment variable in your configuration.


Available Tools (24 Tools)

1. Profile and Runner Matrix (5 tools)

Tool

Parameters

Description

get_enduraw_athlete_profile

None

Athlete biometrics (weight, height, age), Strava connection status, and plan entitlements

get_enduraw_matrix

None

Complete Runner Matrix (global volume, key sessions, cohort percentiles, terrain stats)

get_enduraw_matrix_summary

None

Compact summary of Runner Matrix scores and key indicators

get_enduraw_matrix_history

None

Historical progression and chronological evolution of Runner Matrix scores

get_enduraw_archive_status

None

Status of the activity/FIT archive processing pipeline

2. Athlete Lab and Modeling (3 tools)

Tool

Parameters

Description

get_enduraw_athlete_lab

None

Complete synthesis of the 5 performance pillars, critical speed model, and cohort percentiles (tab=modeling)

get_enduraw_pillar_deep_dive

pillar (enum: endurance, resilience, uphill, downhill, physiology, all)

Granular deep dive into physiological pillars (climbing duration, downhill VDM, cardiac decoupling, resilience calendar)

get_enduraw_performance_profile

None

Critical Speed (Vc, D', benchmark times), radar scores (0-100 on 6 dimensions), and race format ratings

3. Physiological Tests and Evaluations (5 tools)

Tool

Parameters

Description

get_enduraw_physio_tests

None

Physiological evaluation hub: lab test catalogue (Flat, Incline, Lactate) and athlete's real computed benchmarks

get_enduraw_cardio_efficiency

min_speed_kmh (optional), max_speed_kmh (optional)

Cardiac cost of pace across 66 speed bands with median HR, Q25/Q75 dispersion, and post-effort recovery kinetics

get_enduraw_slope_and_cadence

None

Climbing/downhill biomechanics, VAM (m/h) and cadence (spm) broken down by gradient percentage (-25% to +25%)

get_enduraw_cardio_zones

None

Karvonen heart rate zones (Z1 to Z5), threshold BPMs, and total training hours/shares spent in each zone

get_enduraw_records_progression

distance (optional: 1k, 5k, 10k, 15k, semi, 30k, marathon)

Historical personal records and progression curves for benchmark distances with milestones

4. Activities and Deep In-Session Analysis (5 tools)

Tool

Parameters

Description

get_enduraw_activities

limit (default: 10), period (optional: 52w, all, 12w, 4w), analyzed_only (optional boolean)

Synced activities with weather conditions (temperature, wind), duration, distance, and analysis readiness

get_enduraw_activity_analysis

activity_id (number, required)

Deep session analysis: physiological storyline, quality score, heart rate zone distribution, and efficiency metrics

get_enduraw_activities_stats

None

Global activity volume statistics over 52 weeks or lifetime (total distance, observed vs estimated moving time, session count)

get_enduraw_activities_map

None

Complete GPS coordinates array of all activities for personal Heatmap mapping

get_enduraw_analysis_quota

None

Consult current monthly/annual deep-analysis quota and subscription level

5. Race Database, Elevation Profiles and Pacing Plans (4 tools)

Tool

Parameters

Description

get_enduraw_pacing_plans

None

Active and upcoming race pacing and nutrition plans, including available plan credits

get_enduraw_races

query (optional string)

Official race course database (329+ races: UTMB, CCC, Saintelyon, Diagonale des Fous, etc.)

get_enduraw_race_course_profile

race_name (string, required)

Full 3D elevation profile of an official race with D+, D-, min/max altitude, and aid stations count

get_enduraw_race_gpx

race_name (string, required)

Full GPX GeoJSON track and official checkpoints/aid stations with exact distances and cumulative elevation

6. Form and Fatigue (1 tool)

Tool

Parameters

Description

get_enduraw_form_fatigue

days (default: 180), sport (default: 'all')

Official Enduraw Form and Fatigue modeling (CTL, ATL, TSB curves) directly from /auth/api/activities/form-fatigue/. Returns official metrics or subscription requirement status (Starter tier)

7. Notifications (1 tool)

Tool

Parameters

Description

get_enduraw_notifications

today (optional YYYY-MM-DD)

Daily activity notifications, processing alerts, and training updates


Example Prompts

Here are examples of prompts you can use once connected:

  • "What is my Critical Speed (Vc) and D' according to Enduraw, and what are my theoretical times on 5k and 10k?"

  • "Analyze my cardiac cost between 10 km/h and 13.5 km/h. How many beats does each km/h cost me?"

  • "Extract the full 3D course elevation profile and official checkpoints for the CCC."

  • "Show me all 20 aid stations on the UTMB with their exact kilometer markers and cumulative elevation gain."

  • "What is my climbing VAM profile on 10% vs 20% slopes, and at what gradient does my cadence drop?"

  • "Show me my progression curve on the Half-Marathon and Marathon benchmark distances."

  • "What is the breakdown of continuous climbs in my Athlete Lab (<5 min to 20+ min) and my downhill descent speed (VDM)?"


Development

Clone and test locally:

git clone https://github.com/ploa33/enduraw-MCP.git
cd enduraw-MCP
npm install

Copy the example environment file:

cp .env.example .env
# Edit .env with your ENDURAW_ACCESS_TOKEN

Run locally:

npm start

Disclaimer

This project is an independent, community-developed open-source tool built for personal interoperability purposes using the Model Context Protocol. It is not affiliated with, maintained by, authorized by, endorsed by, or sponsored by Enduraw or any of its subsidiaries. All product names, trademarks, and registered trademarks belong to their respective owners.


License

MIT

Available Tools

24 tools
get_enduraw_activitiesB

Fetches recent activities synced to Enduraw with duration, distance, weather conditions (temp, wind), and analysis readiness status. Supports period filtering ('52w', 'all', '12w', '4w') and filtering for analyzed sessions.

Inputs: limit, period, analyzed_only.

get_enduraw_activities_mapA

Fetches geolocation GPS coordinates, start points, and session routes for all recorded activities across the globe (used in the Enduraw Heatmap and geographic explorer).

Inputs: period.

get_enduraw_activities_statsA

Fetches aggregate statistics across all training sessions recorded on Enduraw: total sessions, total distance (km), total time (h), observed vs estimated time breakdown, and time saved.

Inputs: period.

get_enduraw_activity_analysisA

Fetches the deep physiological and pacing analysis of a specific activity by its activityId. Includes storyline, performance claims, quality score, heart rate zones distribution, and weather impact. If the activity has not been analyzed yet due to free tier quotas, returns summary data and quota info.

Inputs: activity_id (required).

get_enduraw_analysis_quotaA

Checks the current user's activity analysis quota on Enduraw (tier, limit, used, remaining, reset date).

Inputs: no named parameters.

get_enduraw_archive_statusB

Checks the processing status of the athlete's uploaded activity archive / FIT history on Enduraw.

Inputs: no named parameters.

get_enduraw_athlete_labB

Directly retrieves the complete Enduraw 'Athlete Lab' / 'Modélisation' (tab=modeling), which is locked for free users on the web app. Unlocks the 5 performance pillars (Uphill continuous climbs, Downhill VDM, Endurance cardiac drift, Resilience 28-day training load calendar, Physiology recovery), Critical Speed profile, and cohort percentiles.

Inputs: no named parameters.

get_enduraw_athlete_profileA

Retrieves the Enduraw athlete profile, settings, biometrics (height, weight, age) and account plan/entitlements.

Inputs: no named parameters.

get_enduraw_cardio_efficiencyB

Retrieves cardiac efficiency vs speed ('Que coute votre allure en battements ?') across 66 speed/pace intervals with median HR, Q25/Q75 dispersion, and post-effort heart rate recovery metrics.

Inputs: max_speed_kmh, min_speed_kmh.

get_enduraw_cardio_zonesA

Retrieves Karvonen heart rate zones (Z1 to Z5), total time and share spent in each zone across training history, resting HR, and max HR.

Inputs: no named parameters.

get_enduraw_form_fatigueB

Fetches the official Enduraw Form & Fatigue modeling (CTL, ATL, TSB, and training load curves) across the athlete's history directly from /auth/api/activities/form-fatigue/. Returns official metrics or subscription requirement status (Starter tier).

Inputs: days, sport.

get_enduraw_matrixA

Fetches the complete Enduraw Runner Matrix (physiological profile: aerobic endurance, climbing, downhill, speed, resilience, cohort percentiles).

Inputs: no named parameters.

get_enduraw_matrix_historyB

Fetches the chronological evolution / history of the athlete's Runner Matrix data points.

Inputs: no named parameters.

get_enduraw_matrix_summaryC

Fetches a summarized view of the athlete's Runner Matrix metrics and current scores.

Inputs: no named parameters.

get_enduraw_notificationsC

Retrieves daily activity notifications and new updates from Enduraw.

Inputs: today.

get_enduraw_pacing_plansA

Retrieves all pacing plans created or in preparation on Enduraw, including available credits.

Inputs: no named parameters.

get_enduraw_performance_profileA

Retrieves deep performance metrics from Enduraw: Critical Speed (Vc, D', benchmark points), radar scores (0-100 on Endurance, Uphill, Downhill, Speed, Physio, Resilience), and performance ratings by race format (short, intermediate, long, very_long).

Inputs: no named parameters.

get_enduraw_physio_testsC

Comprehensive physiological testing hub: retrieves official Enduraw testing protocols (Flat VMA/Vc test, Incline VAM test, Lactate testing), booking availability, cognitive Stroop test status, AND all computed physiological benchmarks extracted from the runner's real history (Critical Speed, Slope VAM curve, Cardiac Efficiency, Post-effort Recovery Kinetics).

Inputs: no named parameters.

get_enduraw_pillar_deep_diveC

Retrieves granular sub-analysis data for any Enduraw physiological pillar: 'endurance' (cardiac decoupling and pace stability by duration), 'resilience' (28-day sustained effort calendar and durability), 'uphill' (continuous climbs and D+ by slope), 'downhill' (continuous descents and VDM), or 'all'.

Inputs: pillar (required).

get_enduraw_race_course_profileB

Retrieves the detailed 3D course elevation profile, distance, total D+, total D-, max altitude, and aid stations count for any of Enduraw's 329+ official races (CCC, UTMB, Sainté, etc.).

Inputs: race_name (required).

get_enduraw_race_gpxB

Retrieves the complete GeoJSON track coordinates and ALL official checkpoints / aid stations (with distance, altitude, and cumulative D+) for a race.

Inputs: race_name (required).

get_enduraw_racesB

Queries the Enduraw official races database (329+ races: CCC, UTMB, Sainté, Diagonale des Fous, etc.) with elevation gain/loss, distances, and date.

Inputs: query.

get_enduraw_records_progressionB

Retrieves the personal records and progression curves computed by Enduraw on standard benchmark distances (1k, 5k, 10k, 15k, Half-Marathon, 30k, Marathon) with historical milestones.

Inputs: distance.

get_enduraw_slope_and_cadenceB

Retrieves VAM (vitesse ascensionnelle in m/h) and running cadence (spm) broken down by slope grade percentage (uphill and downhill). Highlights climbing efficiency and walking/running transition points.

Inputs: no named parameters.

Tool Schema Changelog

Recent tool additions, removals, and schema changes observed during successful MCP inspections.

Loading tool schema changelog…

    TDQS

    B3.3/5.0

    Scored across 24 tools

    Disambiguation3/5

    Most tools target distinct resources, but several physiological-metrics tools overlap heavily: matrix, matrix_summary, matrix_history, performance_profile, athlete_lab, pillar_deep_dive, and physio_tests all expose related profile/benchmark data. Descriptions help, but an agent could plausibly choose the wrong view when asked for 'performance metrics'.

    Naming Consistency5/5

    Every tool follows the same get_enduraw_<snake_case_noun> pattern. The convention is predictable and consistent across all 24 tools.

    Tool Count3/5

    24 tools is at the upper edge for a single server and feels heavy, especially with multiple matrix/performance variants. The broad analytics domain justifies many tools, but some consolidation would reduce surface area.

    Completeness4/5

    Coverage is strong for a read-only running-analytics domain: profile, activities, analysis, physiological pillars, races, course profiles, GPX, form/fatigue, and quotas are all represented. Minor gaps exist around write operations such as creating or editing pacing plans, but core retrieval workflows are covered.

    Maintenance

    ActivityMaintained
    ResponsivenessNo recent community issues

    Related MCP Connectors

    Related MCP Servers

    • A
      license
      B
      quality
      D
      maintenance
      Integrates with the Strava API to allow AI assistants to access fitness data including athlete profiles, activity history, and segment statistics. It enables users to query detailed performance metrics and explore geographic segment data through natural language commands.
      8
      125 npm
      MIT
    • F
      license
      A
      quality
      D
      maintenance
      Enables AI assistants to access and analyze intervals.icu training data including activities, fitness metrics (CTL/ATL/TSB), wellness stats, and calendar events. Supports natural language querying of athletic performance for training insights and workout planning.
      6
      -