Formatting code

Author: pH-7

src/routes/contextRoutes.ts

@@ -1,52 +1,109 @@
-import { FastifyInstance } from 'fastify';
+import { FastifyInstance } from "fastify";
+import cors from "@fastify/cors";
+import axios from "axios";
 
 const name = process.env.MCP_NAME || "MCP Service";
-const description = process.env.MCP_DESCRIPTION || "MCP Server providing model context.";
-const tags = (process.env.MCP_TAGS || "MCP").split(',');
+const description =
+  process.env.MCP_DESCRIPTION || "MCP Server providing model context.";
+const tags = (process.env.MCP_TAGS || "MCP").split(",");
 const email = process.env.MCP_CONTACT_EMAIL || "[email protected]";
 const website = process.env.MCP_CONTACT_WEBSITE || "https://ph7.me";
-const version = process.env.APP_VERSION || '0.1.0';
+const version = process.env.APP_VERSION || "0.1.0";
 
 export async function registerContextRoutes(server: FastifyInstance) {
+  await server.register(cors, { origin: true });
+
+  // Logging middleware
+  server.addHook("onRequest", async (request) => {
+    console.log(`[${new Date().toISOString()}] ${request.method} ${request.url}`);
+  });
+  server.addHook("onError", async (request, reply, error) => {
+    console.error(`[${new Date().toISOString()}] ERROR: ${error.message}`);
+  });
+
   // MCP Discovery Endpoint
-  server.get('/.well-known/model-context', async () => {
+  server.get("/.well-known/model-context", async () => {
     return {
-      "@context": "/.well-known/v1.json",
-      "name": name,
-      "description": description,
-      "version": version,
-      "tags": tags,
-      "contact": {
-        "email": email,
-        "website": website
+      "@context": "/.well-known/v1.json", // Local context reference
+      name: name,
+      description: description,
+      version: version,
+      tags, // ES6 shorthand property name
+      contact: {
+        email: email,
+        website: website,
       },
-      "content_endpoint": "/v1/content"
+      content_endpoint: "/v1/content",
     };
   });
 
   // Available Models Content
-  server.get('/v1/content', async () => {
+  server.get("/v1/content", async () => {
     return [
       {
         title: "Customer Churn Prediction",
-        content: "Predictive model using logistic regression and XGBoost to identify potential customer churn.",
-        tags: ["ML", "churn", "classification", "xgboost"]
+        content:
+          "Predictive model using logistic regression and XGBoost to identify potential customer churn.",
+        tags: ["ML", "churn", "classification", "xgboost"],
       },
       {
         title: "Exploratory Data Analysis Report",
-        content: "Notebook-driven visualisation and summary statistics to uncover key trends and anomalies.",
-        tags: ["EDA", "data analysis", "notebooks"]
+        content:
+          "Notebook-driven visualisation and summary statistics to uncover key trends and anomalies.",
+        tags: ["EDA", "data analysis", "notebooks"],
       },
       {
         title: "Statistical Hypothesis Testing",
-        content: "Support for t-tests, chi-square, ANOVA, and p-value interpretation for decision making.",
-        tags: ["statistics", "hypothesis testing", "inference"]
+        content:
+          "Support for t-tests, chi-square, ANOVA, and p-value interpretation for decision making.",
+        tags: ["statistics", "hypothesis testing", "inference"],
       },
       {
         title: "Time Series Forecasting",
-        content: "Models using ARIMA and Prophet to forecast trends in business KPIs.",
-        tags: ["forecasting", "time series", "prophet"]
-      }
+        content:
+          "Models using ARIMA and Prophet to forecast trends in business KPIs.",
+        tags: ["forecasting", "time series", "prophet"],
+      },
     ];
   });
+
+  // Model Metadata Endpoint
+  server.get("/v1/model/:modelId", async (request) => {
+    const { modelId } = request.params as { modelId: string };
+    if (modelId === "churn") {
+      return {
+        id: "churn",
+        title: "Customer Churn Prediction",
+        input: { type: "json", example: { customerId: "123", features: [/* ... */] } },
+        output: { type: "json", example: { churnProbability: 0.87 } },
+        usage: "POST /v1/predict/churn",
+      };
+    }
+    if (modelId === "eda") {
+      return {
+        id: "eda",
+        title: "Exploratory Data Analysis Report",
+        input: { type: "csv", example: "data.csv" },
+        output: { type: "json", example: { summary: "..." } },
+        usage: "GET /v1/content/eda",
+      };
+    }
+    // ...other models...
+    return { error: "Model not found" };
+  });
+
+  // Strava Activities Integration Endpoint
+  server.post("/v1/strava/activities", async (request, reply) => {
+    const { accessToken, endpoint = "https://www.strava.com/api/v3/athlete/activities", page = 1, per_page = 30 } = request.body as any;
+    try {
+      const response = await axios.get(endpoint, {
+        headers: { Authorization: `Bearer ${accessToken}` },
+        params: { page, per_page }
+      });
+      return response.data;
+    } catch (error: any) {
+      console.error(`[${new Date().toISOString()}] Strava API ERROR:`, error.message);
+      reply.code(500).send({ error: error.message });
+    }
+  });
 }