feat : first OAuth prototype

Author: colinfrisch

README.md

@@ -2,30 +2,14 @@
 
 ### 🏆 MISTRAL MCP HACKATHON WINNER 🏆
 
-@Loucienne
-@leotrois
-@frogens
-@Ulysse6307
-@colinfrisch
+**@Loucienne** </n>
+**@leotrois** </n>
+**@frogens** </n>
+**@Ulysse6307** </n>
+**@colinfrisch** </n>
 
 A Model Context Protocol (MCP) server that provides AI assistants with access to Strava running data, route planning, and weather information. This server enables intelligent running coaching by combining Strava activity analysis with real-time weather data and route generation capabilities.
 
-## Features
-
-### Strava Integration
-- **User Statistics**: Access comprehensive running stats (recent, year-to-date, and all-time totals)
-- **Activity Analysis**: Retrieve detailed information about recent runs including pace, heart rate, and elevation
-- **Performance Visualization**: Generate heart rate and speed charts for activity analysis
-
-### Route Planning
-- **Intelligent Itinerary Creation**: Generate running routes of specified distances from any starting location
-- **Round-trip Route Generation**: Create circular routes that return to the starting point
-- **Google Maps Integration**: Automatically generate Google Maps directions links for easy navigation
-
-### Weather Intelligence
-- **Location-based Forecasting**: Get weather predictions for your usual running areas
-- **Activity-derived Location**: Automatically determine your location from previous run data
-- **Detailed Weather Data**: Access temperature, humidity, precipitation, and wind information
 
 ## Setup
 
@@ -69,16 +53,13 @@ git clone <your-repo-url>
 cd chathletique-mcp
 ```
 
-2. **Install Python dependencies with uv:**
+2. **Install Python dependencies:**
 ```bash
-# Install uv if you don't have it (fast Python package manager)
-curl -LsSf https://astral.sh/uv/install.sh | sh
-
 # Install project dependencies
-uv sync
+pip install -e .
 
 # Install development dependencies (for contributing)
-uv sync --extra dev
+pip install -e .[dev]
 ```
 
 3. **Configure environment:**
@@ -91,97 +72,12 @@ cp .env.example .env
 4. **Set up code quality tools (for contributors):**
 ```bash
 # Install pre-commit hooks for automatic code quality checks
-uv run pre-commit install
+pre-commit install
 
 # Optional: Run pre-commit on all files to check everything
-uv run pre-commit run --all-files
-```
-
-## Usage
-
-### Starting the Server
-
-```bash
-# Using uv (recommended)
-uv run python -m src.strava_mcp.main
-
-# Or activate the environment and run directly
-source .venv/bin/activate  # On Unix/macOS
-python -m src.strava_mcp.main
-```
-
-The server will start on port 3000 and be accessible at `http://localhost:3000/mcp`.
-
-### Testing with MCP Inspector
-
-1. **Install and run the MCP Inspector:**
-```bash
-npx @modelcontextprotocol/inspector
-```
-
-2. **Configure connection:**
-- Transport Type: `Streamable HTTP`
-- URL: `http://127.0.0.1:3000/mcp`
-
-## 📋 Available Tools
-
-### `get_user_stats()`
-Retrieves comprehensive Strava statistics for the authenticated user.
-
-**Returns:** JSON containing recent, year-to-date, and all-time running totals
-
-**Example response:**
-```json
-{
-  "recent_run_totals": {"distance": 50000, "time": 12000, "count": 5},
-  "ytd_run_totals": {"distance": 500000, "time": 120000, "count": 50},
-  "all_run_totals": {"distance": 2000000, "time": 480000, "count": 200}
-}
-```
-
-### `get_last_runs()`
-Fetches detailed information about the user's most recent running activities.
-
-**Returns:** Formatted text with activity details including distance, pace, heart rate, and elevation
-
-**Data includes:**
-- Activity name and date
-- Distance and duration
-- Average and maximum speed
-- Heart rate metrics
-- Elevation gain
-
-### `create_itinerary(starting_place, distance_km)`
-Generates a running route of specified distance starting from a given location.
-
-**Parameters:**
-- `starting_place` (str): Starting location (e.g., "Opéra, Paris")
-- `distance_km` (int): Desired route distance in kilometers
-
-**Returns:** Google Maps directions URL for the generated route
-
-**Example:**
-```python
-# Generate a 10km route starting from central Paris
-create_itinerary("Opéra, Paris", 10)
+pre-commit run --all-files
 ```
 
-### `get_weather_prediction()`
-Provides weather forecast for the user's typical running area based on previous activity locations.
-
-**Returns:** Detailed weather forecast including temperature, humidity, precipitation probability, and wind conditions
-
-**Note:** Requires running `get_last_runs()` first to establish location data.
-
-### `figures_speed_hr_by_activity(number_of_activity, resolution, series_type)`
-Generates heart rate and speed visualization charts for recent activities.
-
-**Parameters:**
-- `number_of_activity` (int): Number of recent activities to analyze
-- `resolution` (str): Data resolution ("high", "medium", "low")
-- `series_type` (str): Series type ("time" or "distance")
-
-**Returns:** List of tuples containing activity names and matplotlib figures
 
 ## Architecture
 
@@ -232,90 +128,6 @@ The project includes automatic quality checks that run before each commit:
 6. **Security scanning**: Detect potential security vulnerabilities
 7. **Spell checking**: Catch typos in code and documentation
 
-### Contributing Guidelines
-
-1. **Install development dependencies:**
-   ```bash
-   uv sync --extra dev
-   uv run pre-commit install
-   ```
-
-2. **Run tests:**
-   ```bash
-   uv run pytest
-   ```
-
-3. **Run quality checks manually:**
-   ```bash
-   # Run all pre-commit hooks
-   uv run pre-commit run --all-files
-
-   # Run specific tools
-   uv run ruff check src/
-   uv run mypy src/
-   ```
-
-4. **Commit your changes:**
-   ```bash
-   git add .
-   git commit -m "Your commit message"
-   # Pre-commit hooks run automatically
-   ```
-
-All code is automatically checked for:
-- **Security vulnerabilities** (Bandit-style checks)
-- **Code style** (PEP 8 compliance)
-- **Import organization** (isort-style)
-- **Type annotations** (MyPy checking)
-- **Documentation quality** (Google docstring format)
-- **Common Python mistakes** (Bugbear checks)
-
-### 🇫🇷 Installation et Pre-commit (French)
-
-**Installation rapide avec uv :**
-```bash
-# 1. Cloner le projet
-git clone <votre-repo-url>
-cd chathletique-mcp
-
-# 2. Installer uv (gestionnaire de paquets Python ultra-rapide)
-curl -LsSf https://astral.sh/uv/install.sh | sh
-
-# 3. Installer les dépendances
-uv sync --extra dev
-
-# 4. Configurer les hooks pre-commit
-uv run pre-commit install
-```
-
-**Comment fonctionne pre-commit :**
-
-Pre-commit est un système qui exécute automatiquement des vérifications de qualité de code **avant chaque commit**.
-
-**Avantages :**
-- ✅ **Code toujours propre** : Impossible de commiter du code mal formaté
-- ✅ **Sécurité automatique** : Détection des vulnérabilités courantes
-- ✅ **Style cohérent** : Formatage automatique selon les standards Google
-- ✅ **Documentation forcée** : Docstrings obligatoires et bien formatées
-- ✅ **Imports organisés** : Tri automatique des imports
-
-**Que se passe-t-il lors d'un commit :**
-1. Vous faites `git commit -m "mon message"`
-2. Pre-commit lance automatiquement tous les outils de qualité
-3. Si des problèmes sont détectés, le commit est **bloqué**
-4. Les outils corrigent automatiquement ce qu'ils peuvent
-5. Vous revérifiez les modifications et recommitez
-
-**Outils inclus :**
-- **Ruff** : Linter ultra-rapide (remplace flake8, isort, black)
-- **MyPy** : Vérification des types Python
-- **Codespell** : Correction des fautes de frappe
-- **Security checks** : Détection de failles de sécurité
-
-## License
-
-This project is licensed under the MIT License - see the LICENSE file for details.
-
 
 ## Use Cases
 

pyproject.toml

@@ -23,19 +23,15 @@ dependencies = [
     # Core MCP framework
     "fastmcp>=2.12.3",
     "pydantic>=2.11.9",
-
     # API clients
     "stravalib>=2.4",
     "openrouteservice>=2.3.3",
     "requests>=2.32.5",
-
     # Data processing
     "numpy>=2.3.3",
     "matplotlib>=3.0.0",
-
     # Geocoding
     "geopy>=2.4.1",
-
     # Environment variables
     "python-dotenv>=1.1.1",
 ]

src/strava_mcp/__init__.py src/chathletique_mcp/__init__.pysrc/strava_mcp/__init__.py renamed to src/chathletique_mcp/__init__.py

@@ -0,0 +1,81 @@
+import os
+
+import requests
+from dotenv import load_dotenv
+from flask import Flask, jsonify, redirect, request
+
+load_dotenv()
+
+app = Flask(__name__)
+app.secret_key = os.getenv("FLASK_SECRET_KEY", "dev_key_123")  # pas important ici
+
+
+STRAVA_CLIENT_ID = os.getenv("STRAVA_CLIENT_ID")
+STRAVA_CLIENT_SECRET = os.getenv("STRAVA_CLIENT_SECRET")
+REDIRECT_URI = "https://gorilla-major-literally.ngrok-free.app/callback"
+
+
+@app.route("/")
+def home():
+    """
+    Redirect to Strava for authentication
+    """
+
+    auth_url = (
+        f"https://www.strava.com/oauth/authorize?"
+        f"client_id={STRAVA_CLIENT_ID}"
+        f"&redirect_uri={REDIRECT_URI}"
+        f"&response_type=code"
+        f"&scope=activity:read_all"
+        f"&approval_prompt=force"
+    )
+    print("Redirecting to Strava:", auth_url)  # Debug
+    return redirect(auth_url)
+
+
+@app.route("/callback")
+def callback():
+    """
+    Callback from Strava after authentication => on récupère le code d'autorisation à partir de l'url de callback
+    """
+    print("Callback received! URL:", request.url)  # Debug
+    print("Args:", request.args)  # Debug: Show all query parameters
+
+    code = request.args.get("code")
+    if not code:
+        return "Error: No 'code' received. Check the Strava app settings.", 400
+
+    print("Authorization code:", code)  # Debug: Print the code
+
+    # Exchange code for token
+    try:
+        response = requests.post(
+            "https://www.strava.com/oauth/token",
+            data={
+                "client_id": STRAVA_CLIENT_ID,
+                "client_secret": STRAVA_CLIENT_SECRET,
+                "code": code,
+                "grant_type": "authorization_code",
+            },
+            timeout=10,
+        )
+        response.raise_for_status()
+        token_data = response.json()
+        print("Token data:", token_data)  # Debug: Print token data
+
+        # Test the token by fetching athlete data
+        athlete_response = requests.get(
+            "https://www.strava.com/api/v3/athlete",
+            headers={"Authorization": f"Bearer {token_data['access_token']}"},
+            timeout=10,
+        )
+        athlete_data = athlete_response.json()
+        return jsonify(
+            {"status": "success", "token": token_data, "athlete": athlete_data}
+        )
+    except Exception as e:
+        return f"Error: {e!s}", 400
+
+
+if __name__ == "__main__":
+    app.run(host="0.0.0.0", port=8000, debug=True)  # noqa