Skip to main content
Glama
Nolio MCP Server

Nolio MCP Server

Nolio MCP Server

An MCP server that gives Claude read-only access to your Nolio training data — planned workouts, completed sessions, metrics, records, and notes.

Prerequisites

Related MCP server: nike-runclub-mcp

Quick Start

1. Clone and install

git clone <repo-url> nolio-mcp
cd nolio-mcp
uv sync

2. Configure client credentials

Create a .env file with your Nolio client credentials:

cp .env.example .env
# Edit .env and fill in NOLIO_CLIENT_ID and NOLIO_CLIENT_SECRET

3. Authenticate with Nolio

Run the built-in OAuth helper to obtain your access and refresh tokens:

uv run nolio-auth

This opens your browser, you log into Nolio, and all credentials (client ID, client secret, access token, refresh token) are saved to ~/nolio-credentials.json.

Important: In your Nolio app settings, set the redirect URI to http://localhost:9876/callback.

4. Add to Claude Code

Since credentials are stored in ~/nolio-credentials.json, you don't need to pass environment variables:

claude mcp add nolio -- uv run --directory /path/to/nolio-mcp mcp run src/nolio_mcp/server.py

You can also re-authenticate at any time from within Claude Code by running /mcp and selecting the authenticate command on the nolio server, or directly via /mcp__nolio__authenticate.

4b. Or add to Claude Desktop

Add to ~/Library/Application Support/Claude/claude_desktop_config.json:

{
  "mcpServers": {
    "nolio": {
      "command": "uv",
      "args": ["run", "--directory", "/path/to/nolio-mcp", "mcp", "run", "src/nolio_mcp/server.py"]
    }
  }
}

No env block needed — the server reads credentials from ~/nolio-credentials.json.

Available Tools

Tool

Description

get_profile

Get your Nolio user profile (id, name)

get_planned_workouts

Retrieve planned workouts from your coach's program

get_workouts

Retrieve completed/logged workouts

get_workout_streams

Get time-series data for a workout (HR, pace, power, etc.)

get_notes

Retrieve notes (injury, availability, illness, rest, goals)

get_metrics

Get a specific metric (sleep, weight, VO2max, HR, etc.)

get_user_metadata

Get user metadata (goals, weight history, HR, FTP)

get_records

Get personal records (power, HR, pace, cadence)

Most tools support optional from_date, to_date (YYYY-MM-DD), and limit parameters.

Available Prompts (Commands)

Prompt

Description

authenticate

Guides you through the OAuth setup to obtain/refresh Nolio API tokens

In Claude Code, run /mcp → select nolio → authenticate, or use /mcp__nolio__authenticate directly.

Development

Testing with the MCP Inspector

uv run mcp dev src/nolio_mcp/server.py

Running unit tests

uv run pytest tests/test_client.py

Running integration tests

Integration tests hit the real Nolio API. Set up credentials first:

cp .test-env.example .test-env
# Fill in .test-env with your Nolio dev credentials
uv run pytest tests/test_integration.py -m integration

API Reference

Full Nolio API documentation: https://github.com/NolioApp/NolioAPI-Documentation/wiki

Available Tools

8 tools
get_metricsA

Get a specific metric entry by its ID. Requires special API authorization.

Args: id: Metric entry ID

Inputs: id (required). Returns: result

get_notesA

Retrieve notes (injury, availability, illness, rest, etc.).

Args: from_date: Start date filter (YYYY-MM-DD) to_date: End date filter (YYYY-MM-DD) limit: Max results to return (default 30) note_type: Filter by type: blessure, dispo, inconfort, goal, indispo, malade, menstruel, repos id: Specific note ID to retrieve

Inputs: id, limit, to_date, from_date, +1 more. Returns: result

get_planned_workoutsB

Retrieve planned workouts from your coach's training program.

Args: from_date: Start date filter (YYYY-MM-DD) to_date: End date filter (YYYY-MM-DD) limit: Max results to return (default 30) id: Specific planned workout ID to retrieve

Inputs: id, limit, to_date, from_date. Returns: result

get_profileA

Get the authenticated user's Nolio profile (id, name).

Inputs: no named parameters. Returns: result

get_recordsA

Get personal records (power, HR, pace, etc.).

Args: cat: Record category — ppr (power), phrr (HR), par (pace), ptr (time), pvar (speed), pcadr (cadence) record_type: Either 'time' or 'distance' from_date: Start date filter (YYYY-MM-DD) to_date: End date filter (YYYY-MM-DD) sports: Comma-separated sport IDs (e.g. '9,23') items: Comma-separated durations in seconds or distances in meters

Inputs: cat (required), record_type (required), items, sports, +2 more. Returns: result

get_user_metadataC

Get user metadata including goals, weight history, HR, sleep, aerobic speed, FTP. Requires special API authorization.

Args: from_date: Start date filter (YYYY-MM-DD) to_date: End date filter (YYYY-MM-DD) limit: Max results to return (default 15)

Inputs: limit, to_date, from_date. Returns: result

get_workoutsA

Retrieve completed/logged workouts.

Args: from_date: Start date filter (YYYY-MM-DD) to_date: End date filter (YYYY-MM-DD) limit: Max results to return (default 30) id: Specific workout ID to retrieve

Inputs: id, limit, to_date, from_date. Returns: result

get_workout_streamsA

Get detailed time-series data for a specific workout (HR, pace, power, cadence, altitude).

Args: id: Workout ID

Inputs: id (required). Returns: result

Tool Schema Changelog

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

Loading tool schema changelog…

    TDQS

    A3.7/5.0

    Scored across 8 tools

    Disambiguation5/5

    Each tool targets a distinct entity (metrics, notes, planned workouts, profile, records, user metadata, workouts, workout streams) with no overlap. Descriptions clearly differentiate their purposes.

    Naming Consistency5/5

    All tools follow a consistent 'get_<noun>' pattern using snake_case, making it easy to predict tool names and understand their actions.

    Tool Count5/5

    8 tools is a well-scoped set for a personal fitness data server, covering key data retrieval needs without being excessive or insufficient.

    Completeness4/5

    The tool set provides comprehensive read access to the platform's data (metrics, notes, workouts, profile, records, metadata, streams). However, it lacks any creation, update, or deletion tools, which may be a gap for agents needing to modify data.

    Related MCP Connectors

    Related MCP Servers

    • F
      license
      Not graded
      quality
      D
      maintenance
      Provides read-only access to your Nike Run Club running activities, including listing runs and retrieving detailed metrics like distance, pace, and heart rate.
      -
    • A
      license
      A
      quality
      D
      maintenance
      Provides AI assistants with read-only access to an athlete's Intervals.icu training data, including activities, wellness metrics, zones, and planned events, for use with MCP clients like ChatGPT and Claude.
      15
      MIT
    • F
      license
      Not graded
      quality
      A
      maintenance
      Enables athletes and coaches to query workouts, build structured intervals, manage calendars, events, health metrics and multi-week plans, and track CTL/ATL/TSB fitness and power PRs through natural language. Uses secure cookie-based authentication that any account can set up in minutes, with no API approval required.
      -